Hi Alexandre,

I do not test with "gdal" . If it is a one shot processing you can try the following "BEAM/VISAT v5.0" , by using reprojection functionalities under "processing/Geometric Operaiton" you can select your output grid and run the process.

On my side it works quite well. Nonetheless results need to be validated.

Enjoy.

sebastien.

Thank you Sebastian,

I didn't knew about BEAM; it will be interesting to explore its functionalities.

Actually I will need to set up a workflow to work with several files. Nevertheless, I could finally perform the transformation with GDAL. The example VRT structure on the link above was wrong, that is why gdalwarp did not recognized the VRT format. If you are interested in an option with GDAL, I'm attaching the correct VRT structures to use for this procedure, with data for projection of the test file above (they can be edited as a text file).

With all files (includinhg the HDF5) on the same folder, the gdal command line would be:

gdalwarp -geoloc -r cubicspline -s_srs "epsg:4326" -t_srs "+proj=longlat +ellps=WGS84 +datum=WGS84 +no_defs" -tr 1 1.25 TOMS-Earth-Probe_L2-TOMSEPL2_2000m1231t2303-o24142_v001-2004m0424t164259_1.vrt test.tif

The output resolution of 1ºx1.25º is to mach the commom TOMS L3 data.

Best regards,

I use the AMSR-2 brightnes temperature data, and use your vrt format, but it said "HDF5:GW1AM2_201401010714_011A_L1SGRTBR_2210210.h5://"Brightness Temperature (89.0GHz-A， H)"" does not exist in the file system and is not recogised as a supported datasetname. Actually I have tested "gdalinfo GW1AM2_201401010714_011A_L1SGRTBR_2210210.h5". It is OK. Maybe there are something wrong with "Brightness Temperature (89.0GHz-A， H)". I used the data from AMSR-2 1LB brightness temperature, Do you know How can I know how to use it? Thanks!

 Sorry ,it is "HDF5:GW1AM2_201401010714_011A_L1SGRTBR_2210210.h5://"Brightness_Temperature_(89.0GHz-A， H)", but it has the same error.

Hello Qunou,

To get the correct dataset name you have to explore your file using basic hdf5 tools, like h5dump (https://www.hdfgroup.org/HDF5/doc/RM/Tools.html#Tools-Dump).

An example would be:

h5dump -n GW1AM2_201401010714_011A_L1SGRTBR_2210210.h5 (if the file is in the same folder you are running the command, otherwise provide a full or relative path). Check to see if the path inside the file to your data is correct.

Separate runs with the options -A (attributes) and -H (headers) may also be interesting for you.

Hi Alexandre Castagna,

Thank you so much. I have read the h5dump, and I want to know where I can type in the command line. Are there some software? Where I can download it?

Hi Qunou,

The utilities can be downloaded for linux with standard application managers (like: "apt-get install hdf5-tools"). In the website i sent you, they release compiled tools and libraries for Windows, but i dont have experience with those. In a linux system, just run the command on a terminal (equivalent to the command prompt on windows). Possibly it would be something similar in Windows.

Hi  Alexandre Castagna,

Thank you for your advise. I have run it successfully. I got the results as follows.

HDF5 "e:\GW1AM2_201401010714_011A_L1SGRTBR_2210210.h5" {

FILE_CONTENTS {

group /

dataset /Area Mean Height

dataset /Attitude Data

dataset /Brightness Temperature (original,89GHz-A,H)

dataset /Brightness Temperature (original,89GHz-A,V)

dataset /Brightness Temperature (original,89GHz-B,H)

dataset /Brightness Temperature (original,89GHz-B,V)

dataset /Brightness Temperature (res06,10.7GHz,H)

dataset /Brightness Temperature (res06,10.7GHz,V)

dataset /Brightness Temperature (res06,18.7GHz,H)

dataset /Brightness Temperature (res06,18.7GHz,V)

dataset /Brightness Temperature (res06,23.8GHz,H)

dataset /Brightness Temperature (res06,23.8GHz,V)

dataset /Brightness Temperature (res06,36.5GHz,H)

dataset /Brightness Temperature (res06,36.5GHz,V)

dataset /Brightness Temperature (res06,6.9GHz,H)

dataset /Brightness Temperature (res06,6.9GHz,V)

dataset /Brightness Temperature (res06,7.3GHz,H)

dataset /Brightness Temperature (res06,7.3GHz,V)

dataset /Brightness Temperature (res06,89.0GHz,H)

dataset /Brightness Temperature (res06,89.0GHz,V)

dataset /Brightness Temperature (res10,10.7GHz,H)

dataset /Brightness Temperature (res10,10.7GHz,V)

dataset /Brightness Temperature (res10,18.7GHz,H)

dataset /Brightness Temperature (res10,18.7GHz,V)

dataset /Brightness Temperature (res10,23.8GHz,H)

dataset /Brightness Temperature (res10,23.8GHz,V)

dataset /Brightness Temperature (res10,36.5GHz,H)

dataset /Brightness Temperature (res10,36.5GHz,V)

dataset /Brightness Temperature (res10,89.0GHz,H)

dataset /Brightness Temperature (res10,89.0GHz,V)

dataset /Brightness Temperature (res23,18.7GHz,H)

dataset /Brightness Temperature (res23,18.7GHz,V)

dataset /Brightness Temperature (res23,23.8GHz,H)

dataset /Brightness Temperature (res23,23.8GHz,V)

dataset /Brightness Temperature (res23,36.5GHz,H)

dataset /Brightness Temperature (res23,36.5GHz,V)

dataset /Brightness Temperature (res23,89.0GHz,H)

dataset /Brightness Temperature (res23,89.0GHz,V)

dataset /Brightness Temperature (res36,36.5GHz,H)

dataset /Brightness Temperature (res36,36.5GHz,V)

dataset /Brightness Temperature (res36,89.0GHz,H)

dataset /Brightness Temperature (res36,89.0GHz,V)

dataset /Earth Azimuth

dataset /Earth Incidence

dataset /Land_Ocean Flag 6 to 36

dataset /Land_Ocean Flag 89

dataset /Latitude of Observation Point for 89A

dataset /Latitude of Observation Point for 89B

dataset /Longitude of Observation Point for 89A

dataset /Longitude of Observation Point for 89B

dataset /Navigation Data

dataset /Pixel Data Quality 6 to 36

dataset /Pixel Data Quality 89

dataset /Position in Orbit

dataset /Scan Data Quality

dataset /Scan Time

dataset /Sun Azimuth

dataset /Sun Elevation

}

}

So I guess it is correct as attached files. But it still says that "it does not exist in the file system, and is not recognised as a supported dataset name".

Hello Qunou,

When I open your file, the data set is written as: Brightness_Temperature_(original,89GHz-A£¬H)

when it should be: Brightness_Temperature_(original,89GHz-A,H)

But in any case, have you tested the projection with the file I sent in the post, using the correct vrt files I sent later?

I just used this procedure once, so is difficult to guide you any further, because after that I started to use R for the "projection". The method work like that: open the data set in R, as well as the Lat and Long coordinates for the center of each pixel, create a spatial point dataframe representing the original pixels as points (their centroids), and rasterize it using a mask with the desired properties (pixel size). The limitation of this approach is that it does not consider the overlap of the original pixel with two or more new pixels, it will attribute the value of the original pixel just to the new pixel were the centroid is located. I used that because this simple projection (without weighted means based on the amount of overlap) was used by the scientific team of the product I was working with. So if it was good for them, it was good for me (I don't have any background in geodesy anyway). Tell me if its enough for you, and then I can give you some help - if you could not find any help for the more appropriate procedure, with this approximation at least you can do the projection you need.

But don't let to consider BEAM, as Sebastian has pointed out. I don't have experience with that, but certainly should it should have a large community of users.

Hi Alexandre,

Have you used  "BEAM/VISAT v5.0" to transform the projection of brightness temeprature of ASMR-2 data?  It is special, and its location information is hidden in one channel of 89GHz. So it is not easy to transform it as usual.