30m data not available via cropscape for California years 2007 and 2008

[qjhart]

_Sent to HQ_RDDGIB@nass.usda.gov @ 2015-05-12

This is the data contact for the CDL data, as per, http://www.nass.usda.gov/research/Cropland/metadata/metadata_ca07.htm.

I have tried multiple methods to retrieve California's CDL at 30m for 2007, which according to the metadata should be available. Going through Cropscape (http://nassgeodata.gmu.edu/CropScape/) only makes available the 56m data product. I've also tried the caches at: nassgeodata.gmu.edu/nass_data_cache/byfips. And going through http://datagateway.nrcs.usda.gov/ doesn't allow the specification of a date for the CDL.

The same problem exists for years 2008 and 2009 via the above access points, however those layers are available via the national map downloads, though somewhat inconveniently.

Is there as location where I can retrieve this data?

And Later that day...

Thanks for the response. I understand that. You see you refer to the 30m for 2007 and 2008, but you cannot get those from cropscape: http://nassgeodata.gmu.edu/CropScape/ from what I can tell. I'm still trying to find a way to download the 2007 and 2008 30m data product via cropscape, if possible. I have since discovered that the datagateway link, http://datagateway.nrcs.usda.gov/ does respond with every CDL year, but it's in a different projection, and so for consistency, it would be nice if there was a link somewhere to the 2007 and 2008 CA 30m data in the more standard Albers Conic Projection. I guess what I'm saying is that IMO, the Cropscape geotiff prodcuts, like: http://nassgeodata.gmu.edu/nass_data_cache/byfips/CDL_2007_06.zip are the nicest download method but only exist for the 56m data for 2007 and 2008.

[qjhart]

_Response from HQ_RDDGIB@nass.usda.gov @ 2015-05-12

I just downloaded the 2007 and 2008 California CDLs and I also got 56m data. I will contact the folks who manage our data on the server and figure out what's going on. I will get back to you once I know more.

[qjhart]

While waiting, my current method is to pull the 2008 data from the national dataset. and pull the 2007 from http://datagateway.nrcs.usda.gov/.

The national pull is simple, but time consuming, download the 2008 national data, and upload the 30m data then pull over a subset for California. Within the conterminous_us mapset, I just did:

g.region rast=CDL_2010_06;
r.mapcalc 2008_30m_cdls_06=2008_30m_cdls;
g.remove 2008_30m_cdls;

The data from http://datagateway.nrcs.usda.gov/ is more problematic. For this, I downloaded the complete set of CDL data, which comes as one blob from the data gateway. I then created a UTM11 location, and imported just the 2007 and 2008 data (I use the 2008 data to look at the size of errors in the transform). This can't be easily automated, just go to website, and get the CDL data for California.

Then do:

# Import from data gateway glob
g.mapset mapset=cdl location=utm11
r.in.gdal input=cdl_30m_r_ca_2008_utm11.tif output=cdl_30m_r_ca_2008
# Now project to our standard
g.mapset mapset=cdl location=conterminous_us
g.region rast=2010_CDL_06
r.proj input=cdl_30m_r_ca_2008 location=utm11

You can see the difference in sizes:

Input:
Cols: 29767 (29767)
Rows: 35841 (35841)
North: 4674871.000000 (4674871.000000)
South: 3599641.000000 (3599641.000000)
West: -128849.000000 (-128849.000000)
East: 764161.000000 (764161.000000)
EW-res: 30.000000
NS-res: 30.000000

Output:
Cols: 23648 (23648)
Rows: 40345 (40345)
North: 2452725.000000 (2452725.000000)
South: 1242375.000000 (1242375.000000)
West: -2356095.000000 (-2356095.000000)
East: -1646655.000000 (-1646655.000000)
EW-res: 30.000000
NS-res: 30.000000