H David,

Unfortunately the MGET tool you mention can only handle SDSes that have 2 dimensions. If you try the tool on an SDS with 4 dimensions, it will report an error.

If you are skilled at Python programming, you could write a script that extracts the data as a series of 2 dimensional slices and then writes them out as rasters:

1.    Use the pyhdf module (http://pysclint.sourceforge.net/pyhdf/) to extract slices as 2D numpy arrays.

2.    From the GeoEco.DataManagement.ArcGISRasters module import the ArcGISRaster class and then call the FromNumpyArray method to write a raster for each 2D numpy array.

If you send me a link to the HDF that you're trying to process, I may be able to write a short script that demonstrates this technique. However I do not have time to provide you complete technical support. If you are comfortable with computer programming then my example may be helpful. But if you are not a programmer, then it will be too difficult for you to rely on my help.

For MGET 0.8, we are redeveloping the conversion and sampling tools to work much faster, to have fewer dependencies on ArcGIS, and to support 3D and 4D datasets. A big goal is to simplify the sampling of time series data, so that we have a single tool that can implement the procedure described by this example: http://code.env.duke.edu/projects/mget/wiki/Sampling%20time%20series%20rasters. Another goal is to support direct sampling of data in HDF and NetCDF formats, and accessible via OPeNDAP. Currently, you must convert these data to ArcGIS format before doing the sampling, making it very slow. The new tools will not be available for a few months. If you can afford to wait until then, it might be your easiest option.

Jason

Hi, my name is David Fonseca.
I am chilean student of engineering.

Can MGET convert SDS in HDF to Raster if the SDS have 4 dimension???

How??

Thanks..

Ingresa a tu Hotmail desde tu Messenger. ¡Windows Live hace tu vida más simple!