Download MODIS data

[mehdidar...@gmail.com]

Dear David and all,

I encountered a problem to download the MODIS data. Since my data is related to the 2017 and OSCAR website does not have this period and work as well, I decided to download my MODIS data manually. I downloaded some data but I noticed that there are several sub-band in the MOD05L2 product. I want to know that which product must be exactly selected.
I assume one the following '' water_vapor_infrared_mod05 (1km) and water_vapor_near_infrared _mod05 (5km)'' must be the proper product?

As the format of MODIS data is HDF and Train needs GRD format, do you have any suggestion how to convert the data from HDF to GRD?

Best,
Mehdi   

[David Bekaert]

hi,

Correct in terms of the products.

You could use gdal on how to convert to tiff or grd.

GDAL can read from hdf5 directly.

you would need to specify the variable name and use gdal_translate -of GMT or use -of tiff.

Note you might still need to apply the cloud mask which is also a layer inthe hdf5 file.
Would just format the data you downloaded manually such it looks like a regular OSCAR downloaded file.

If you do that then you should be able to use the other steps from TRAIN to calculate the delay.

Cheers,

David

--
You received this message because you are subscribed to the Google Groups "TRAIN - Toolbox for Reducing Atmospheric InSAR Noise" group.
To unsubscribe from this group and stop receiving emails from it, send an email to TRAIN_support+unsubscribe@googlegroups.com.
To post to this group, send email to TRAIN_...@googlegroups.com.
Visit this group at https://groups.google.com/group/TRAIN_support.
To view this discussion on the web visit https://groups.google.com/d/msgid/TRAIN_support/6833250d-6482-44d3-89e3-6efffed28f3d%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[mehdidar...@gmail.com]

Dear David,

Thanks for the helpful guidance.
I converted hdf files to grd format and after runing the aps_modis(3,3), I faced the following error:

Error in aps_modis_InSAR (line 210)
            d_modis_no_interp(:,k) = xyz_output_nointerp(:,3);

Error in aps_modis (line 116)
    aps_modis_InSAR

When I checked the output files of the aps_modis(2,2) step (i.e. *_ZWD_gauss.xyz, *_ZWD_nointrep.xyz, and *_ZWD_surf.xyz), I noticed that the files are empty (0 bytes)!

I checked the grd files created by GDAL using the grdinfo command and it shows:

MOD05_20171101.grd: x_min: 0 x_max: 1354 x_inc: 1 name: meters nx: 1354
MOD05_20171101.grd: y_min: 2030 y_max: 0 y_inc: -1 name: meters ny: 2030
MOD05_20171101.grd: z_min: 1 z_max: 8142 name: meters
MOD05_20171101.grd: scale_factor: 1 add_offset: 0

It seems taht the files do not have a proper georeference coordinate (x, y min/max). I do not know the problem is related to the gdal-converted data or MODIS delay processing in Train? I appreciate if you could give me a hint that what could be the problem in my work.
I attached a grd file converted by GDAL.

Regards,
Mehdi

On Friday, February 9, 2018 at 5:26:45 PM UTC+1, David Bekaert wrote:

hi,

Correct in terms of the products.

You could use gdal on how to convert to tiff or grd.

GDAL can read from hdf5 directly.

you would need to specify the variable name and use gdal_translate -of GMT or use -of tiff.

Note you might still need to apply the cloud mask which is also a layer inthe hdf5 file.
Would just format the data you downloaded manually such it looks like a regular OSCAR downloaded file.

If you do that then you should be able to use the other steps from TRAIN to calculate the delay.

Cheers,

David

On 9 February 2018 at 05:46, <mehdidar...@gmail.com> wrote:

Dear David and all,

I encountered a problem to download the MODIS data. Since my data is related to the 2017 and OSCAR website does not have this period and work as well, I decided to download my MODIS data manually. I downloaded some data but I noticed that there are several sub-band in the MOD05L2 product. I want to know that which product must be exactly selected.
I assume one the following '' water_vapor_infrared_mod05 (1km) and water_vapor_near_infrared _mod05 (5km)'' must be the proper product?

As the format of MODIS data is HDF and Train needs GRD format, do you have any suggestion how to convert the data from HDF to GRD?

Best,
Mehdi   

--
You received this message because you are subscribed to the Google Groups "TRAIN - Toolbox for Reducing Atmospheric InSAR Noise" group.

To unsubscribe from this group and stop receiving emails from it, send an email to TRAIN_suppor...@googlegroups.com.

[mehdidar...@gmail.com]

Hi David,

I could solved previous problem regarding the geo-coordinate info in my data-translated by gdal as follows:

gdalwrap -overwright Water_Vapor_Near_Infrared_name_band.hdf  -of netCDF MODIS_yyyymmdd.nc
gdal_translate MODIS_yyyymmdd.nc -of  GMT MODIS_yyyymmdd.grd

then I estimated the delay over the Alpine region using eighteen S-1 data. In 21 August date, there is some intense changes (fig.1) but when I removed the this data from the delay estimation some changes appeared in 8 Oct. data (fig.2), and after removing this data, some changes appeared in the 15 August data (fig.3). In this way, after removing the 15 August data from the delay estimation, the changes was observed on the rest of the data (fig.4).It seems there is sth wrong with these three dates. By the way, I did not apply the cloud mask band. Concerning the initial results, I would be appreciated if you could help me to interpret the results and guide me how to figure out the problem.

First, the range of the color bar is weird in figures (i.e. -2.4e+11 - 1e+11)? I am not enough familiar with MODIS data but the range color bar on the quick look image of water band data (MODIS) is nearly 0-8 but the range pixel values in the hdf file is nearly 0-8000 (maybe factor of 1000). I do not know to what extent applying the Cloud Mask band (8 bit) could affect the results and what criterion should be chosen for the threshold selection to mask the data? I attached the water vapor and cloud mask MODIS data corresponding to the SAR data (fig.5) and the linear delay estimated ('a-l') (fig.6) as well. 

Thanks for any hint in advance.
Regards,
Mehdi

[mehdidar...@gmail.com]

Sorry but it seems that the fig. 5 is not shown properly in the web browser, so I re-attached it here with a different format.

[12271...@qq.com]

hi,
   I recently wanted to download MODIS data manually and use it. I would like to ask you a few questions: Can you write hdf to grd code?



在 2018年2月28日星期三 UTC+8下午7:23:35，mehdidar...@gmail.com写道：

[iapatric...@gmail.com]

Hi Mehdi,

I have the same issue you had, do you think you could help me? I did a transformation using R and gdal-translate but it seems it doesn't work when running step 2 and have no idea what I did wrong.

Cheers!

[reyhanaz...@mail.ugm.ac.id]

Dear Mehdi,

Do you write Gdalwarp correctly? Because I can't find -overwrigth on gdal and I also have problems like you. Thank you

Best regards,

Reyhan