Improve ASTER latitude, longitude grid calculation #364
Conversation
| @@ -919,3 +950,15 @@ def cloudtopheight_IR(bt, cloudmask, latitude, month, method="modis"): | |||
| bt_clear_avg = np.nanmean(bt[mask_clear]) | |||
|
|
|||
| return (bt_clear_avg - bt_cloudy) / lapserate | |||
|
|
|||
| def geocentric2geodetic(latitude): | |||
There was a problem hiding this comment.
There is a function for that in typhon.geodesy. I am not sure if it does the same, but you could check that.
There was a problem hiding this comment.
True! However, the conversion given here is from the ASTER user guide. Might be an approximation, but as it's stated there I'd like to stay consistent within the ASTER L1B data. The typhon.geodesy seems more complicated (might be more precise?)
There was a problem hiding this comment.
Fair enough! You may mention that in the docstring or define the function as a staticmethod of the Aster image class. But having it in the typhon.cloudmask.aster may already sufficient ;)
typhon/cloudmask/aster.py
Outdated
| "TIR": (14, 15) | ||
| } | ||
|
|
||
| latstr = (f"HDF4_SDS:UNKNOWN:{self.filename}:{sdsidx[self.channel2sensor[channel]][0]}") |
There was a problem hiding this comment.
Is there a reason for the parantheses around the f-string?
There was a problem hiding this comment.
No, but the string is too long :) I'll fix it
typhon/cloudmask/aster.py
Outdated
| '5', '6', '7', '8', '9', '10', '11', '12', '13', '14'. | ||
|
|
||
| Returns: | ||
| tuple(ndarray, ndarray): latitude, longitude |
There was a problem hiding this comment.
| tuple(ndarray, ndarray): latitude, longitude | |
| ndarray, ndarray: latitude, longitude |
| latidx = np.arange(11) * imagedimension.lat / 10 | ||
| lonidx = np.arange(11) * imagedimension.lon / 10 |
There was a problem hiding this comment.
Are the image dimensions always dividable by ten? Otherwise you could use an integer division x // 10 to ensure an int result
There was a problem hiding this comment.
The ASTER L1B user guide specifies geolocation boxes, their sizes, and the corresponding lattice points (which don't change).latidx and lonidx are those lattice points and are only needed for defining the interpolation function and not necessarily need to be int.
ASTER subsensor information and channel information are connected. The derivation of channel information changed, but not accessing the variable.
* `geocentric2geodetic`: transform latitude values from geocentric to geodetic values. * `read_coordinates`: reada reduced latitude, longitude grid from ASTER HDF file.
* `get_latlon_grid`: instead of a linear interpolation between the scene corner coordinates, the more detailed geolocation information from sds in ASTER HDF files is used and linearly interpolated in between the 11 x 11 defined grid points per image.
tmieslinger
force-pushed
the
change_latlon_calc
branch
from
78880a8
to
1aa1fef
Compare
No description provided.