Help Topics/Data Sets/NOAA

How do I import SNODAS data into ArcGIS?

Kara
posted this on July 7, 2011, 13:34

Question:

How do I import SNODAS data into ArcGIS?

Answer:

Follow the steps below (which are also linked to from the data set documentation.)

These steps will work for all 8 SNODAS variables.  In the below example I will be importing snow depth data from February 1, 2007.

filename= us_ssmv11036tS__T0001TTNATS2007020105HP001.dat

1. Change the "filename" extension from .dat to .bil.

2. Create the following header file and give it the same name as "filename" but with a .hdr extentsion. Remember to place in the same directory as "filename.bil"

byteorder M
layout bil
nbands 1
nbits 16
ncols 6935
nrows 3351
ulxmap -124.729583333331703
ulymap 52.871249516804028
xdim 0.00833333333
ydim 0.00833333333

Note: this header file came from page 12 of Barrett, A. 2003. National Operational Hydrologic Remote Sensing Center Snow Data Assimilation System (SNODAS) Products at NSIDC. NSIDC Special Report 11. Boulder, CO, USA: National Snow and Ice Data Center. 19 pp.  The header information may be different for individual files. See the header file provided with each data file during FTP download for specific information.

3. Start ArcMap and open ArcToolbox. Select Conversion Tools > To Raster > Raster to Other Format. The Raster to Other Format window will appear. In the Input Raster box, browse and select filename with a .bil extension (Refer to Step 1.). In the Output Workspace box, select the location to store the new data set. In the Output Format box, select the format to create the new data set; this can be one of three formats: ESRI Grid, ERDAS Imagine, or TIFF. Click OK. This new grid will be renamed filename1=snowdepth020107

4. For this next step the whole file path name must be included (ie if filename1 is located in D directory the file path name is D:\filename1). Since SNODAS is16 bit- signed integer data, data needs to be formatted that way. In ArcGIS open up the ArcToolBox → Go to Spatial Analyst Tools → Then go to Map Algebra → and click on Multi Output Map Algebra. In the new input box, type the following command in the Map Algebra Expression box:

D:\filename1conv = con(D:\filename1>= 32768, D:\filename1- 65536, D:\ filename1)

Click OK.
***Make sure that the syntax for the expression is followed exactly how it is displayed here (ie space between numbers and/or file name) other wise the conversion won’t work or be incorrect).

Note: Step 4 can be skipped if you have  Arc 10 and use this header file instead:

nrows 3351
ncols 6935
nbands 1
nbits 16
pixeltype signedint
byteorder M
layout bil
ulxmap -124.729583333331703
ulymap 52.871249516804028
xdim 0.00833333333
ydim 0.00833333333

If you have spatially subset data through our Polaris interface and output your data in the native binary format, the header file will again be different. Each data file is delivered with a .dat.sub.txt file that you will need to reference. Your header file should contain:

byteorder I
layout bil
nbands 1
nbits 16
ncols xxxx   (corresponds to Image Width in .dat.sub.txt file)
nrows xxxx (corresponds to Image Height in .dat.sub.txt file)
ulxmap xxxx (corresponds to Upper Left Corner X Coordinate in .dat.sub.txt file)
ulymap xxxx (corresponds to Upper Left Corner Y Coordinate in .dat.sub.txt file)
xdim 0.00833333333
ydim 0.00833333333

 

 
Topic is closed for comments