search

GEUS-Glaciology-and-Climate / pypromice

Process AWS data from L0 (raw logger) through Lx (end user)
https://pypromice.readthedocs.io
GNU General Public License v2.0
12 stars 4 forks source link

Revisit tilt correction of SR or use the correction by Wang et al (TC, 2016) #28

Open mankoff opened 3 years ago

mankoff commented 3 years ago

https://github.com/GEUS-PROMICE/PROMICE-AWS-processing/blob/a1508a6b06dc1ce749b0fa95c43a7879cb0993f1/IDL/AWSdataprocessing_v3.pro#L14

PennyHow commented 1 year ago

The SR is currently corrected for sensitivity to air temperature only.

https://github.com/GEUS-Glaciology-and-Climate/pypromice/blob/2331c02e1e121648c06dd054d370f0b99a6a8f6d/src/pypromice/L0toL1.py#L58-L59

If someone has the full reference to find this paper then we can look at implementing a correction based on station tilt.

citterio commented 1 year ago

If someone has the full reference to find this paper then we can look at implementing a correction based on station tilt.

Quite sure https://doi.org/10.5194/tc-10-727-2016 (for shortwave radiation). We do have the tilt and, increasingly, azimuth measurements, so they seem preferable to the Wang method except if those sensors fail, and perhaps for identifying problems like misaligned sensors.

PennyHow commented 1 year ago

Thanks Michele. I see that the paper nicely documents the radiometer tilt correction.

For reference, the radiometer measurements are currently adjusted for tilt here: https://github.com/GEUS-Glaciology-and-Climate/pypromice/blob/f3c8050a4b2a65d7bc80249a3f864e3bcf19d31e/src/pypromice/L1toL2.py#L63-L79 This resource is noted in the documentation for calculating the correction factor (calcCorrectionFactor)

I'll take a look at this and decide if the method described in this paper is a better (or worse) alternative.

patrickjwright commented 1 year ago

I haven't looked at the details yet, but wondering.... is it possible to apply tilt correction methods for radiation to sonic data? The radiation corrections have to do with variation in direct/diffuse fields, the sensor potentially seeing upwelling in the downwelling field, etc... whereas if a sonic is out of level, the path-length (and transit time) becomes longer for the emitted sonic signal. These seem like quite different problems?

BaptisteVandecrux commented 1 year ago

As Patrick says, this post mixes two things: shortwave radiation measurements and SR50 measurements. I think Dirk's comment was originally for shortwave (SR) measurement only.

For radiation the current tilt correction is described in Robert's ESSD paper. Changing to Wang et al. (2016) should be done only if it significantly improves dsr and usr correction and the impact of this change should be documented.

The only part of Wang et al. (2016) we could use already is this RIGB algorithm which estimates tilt from radiation measurements and then applies the tilt correction. This could be used for stations or years when direct measurement of tilt is not available. I don't know how often tilt-meter failure happened in the past.

The SR50 has a field of view of 30deg, it will pick up the closest object within it. So it should still pick up the vertical distance to the surface in case of minor tilt. Worst case, it might detect the station itself, but then there is no correction that can be applied to fix this issue.