How to process modis hdf files already available on disk using MODIStsp #142
Comments
|
Hi, you can find some instructions for offline processing at: https://ropensci.github.io/MODIStsp/articles/faq.html#working-with-already-downloaded-hdf-files Let us know if you have problems! |
|
I placed all the HDF files (h23-27, V04-06, 15files) in 'G:\VIP\GWR\MCD18A1\HDF',like this
Thanks a lot! |
|
It didn't do the mosaicing and reprojecting work. I do not know why? |
|
could you try changing the start / end dates to 2011-01-01 and 2011-12-31 ? |
|
I changed the start/end dates to 2011-01-01 and 2011-12-31, it also didn't work. |
|
Sorry it did not work. Was just a wild guess. I'll look into this more in detail, but considee that i will be on holiday for the next couple of week, so it will take some time. |
|
Thanks a lot! waiting for you. |
|
Was this ever resolved? I am experiencing the same problem.
(rsession.exe:12596): Gtk-WARNING **: Failed to set text from markup due to error parsing markup: Error on line 1 char 87: Element 'markup' was closed, but the currently open element is 'span' (rsession.exe:12596): Gtk-WARNING **: Failed to set text from markup due to error parsing markup: Error on line 1 char 87: Element 'markup' was closed, but the currently open element is 'span' (rsession.exe:12596): Gtk-WARNING **: Failed to set text from markup due to error parsing markup: Error on line 1 char 91: Element 'markup' was closed, but the currently open element is 'span' (rsession.exe:12596): Gtk-WARNING **: Failed to set text from markup due to error parsing markup: Error on line 1 char 91: Element 'markup' was closed, but the currently open element is 'span'
|
|
Hi, @webb-e so, you have some HDFs in D:/MODIS_burns_originalHDF and you want to run the processing offline, is that correct? What exactly happens when you try? Lorenzo |
|
That is correct. MODIStsp creates a folder called "Burned_Monthly_500m_v6" in the folder where I asked it to save the processed tiff files, but the folder is empty. The R output is above. I should note that the program only runs for a few seconds before ending, so I don't think it starts processing anything. |
|
Yeah, that's probable. May I ask you to save your options to JSON (using the save options button in the GUI) , open the JSON file in a text editor and cut and paste it here ? That way, I can more easily replicate your analysis. |
|
{ |
|
sorry for the hiatus. This should be now fixed on the "devel" branch. Could you test it? Lorenzo |
|
Hmm. So, this worked for the files in 2001 and 2002, but not for any of the other files. Here is a warning I got:
(but those warnings might be because there were no burns at that date? the file exists though)
|
|
That's strange. I tested it on 2010 data and it worked. Maybe you have some corrupted / incomplete HDF.
(It's a single command - no carriage return) |
|
I tried it on only 2010 data with no luck. Here is the output for my entire session I checked a couple of the hdf files in the way you suggested and the console printed out their metatdata (so this means they are not corrupt?) |
|
Also, I just ran the data just for 2001-2003 and it did not work. I suspect last time I had some old files in the folder, so in fact when I ran the program yesterday it did not work for any years (based on their file creation date, the 2001-2003 files do not look like they were processed yesterday when I rant the program) |
|
I wonder if this could be something similar to #114.... On Windows, there seems to be a "limit" to how many tiles we are able to "stitch together". To verify this, could you to :
but with a different name for the vrt file.
"start_x": 20, |
|
(I used the same vrt file as earlier since I had not closed my R session. I also searched for mstp_temp I tried a smaller extent and it does appear to have worked (there are output files but I have not checked them yet). I used this: However, I will note that earlier I directly downloaded the data and MODIStsp processed the large extent. Actually, it did crash when I used: I'm not sure why, but when I changed the extent to x:26 instead of x:27, it worked. The plan was just to add that last tile in after the fact. The reason that I don't have the processed data from the first download is because when I went to download that last tile (h27, v3), it replaced all of my previously processed files. I'm new to this and didn't know that was going to happen, so now I am just trying to process the files that I originally downloaded. However, it seems like the fastest option might be to just re-download them, since processing worked that way. |
|
Ok. Then I'd suggest you to try to process 4/26 1/3 using "http" download option. Consider that even in that case, you will not need to re-download the HDFs if you do not change the HDF output folder. MODIstsp will just check to verify if you already have all the HDFs for the specified spatial/temporal selection. If it works, good. otherwise, I suggest you to try to do two separate processings (e.g, 4-->12 and 13-->26), (with different output folders) and see if by reducing the number of tiles you can avoid the problem. If it works, you can then merge the results afterward. |
|
Closing here. Keeping open in #114 |
Hi, I need to download MCD18A1 radiation data from May to Nov. in 2011. However, MCD18A1 data can only be downloaded before 2009 in the https://e4ftl01.cr.usgs.gov/MOTA/MCD18A1.006/. I found the MCD18A1 data in 2011 can be download in the https://ladsweb.modaps.eosdis.nasa.gov/search/order/4/MCD18A1--6. Thus, I want to know how to process MODIS HDF files already available on disk using MODIStsp.
Thanks a lot!
The text was updated successfully, but these errors were encountered: