Source:         Princeton University
Name:           Princeton Global Forcings
Period:         Jan-1948 - Dec-2000
Times:          Monthly
Coverage:       Global
Resolution:     1.0-degree
Grid:           Latitude-longitude
Type:           Observational-Reanalysis hybrid
Related Sites:	http://hydrology.princeton.edu/data/index.html
Publications:   Sheffield, J., G. Goteti, and E. F. Wood, Development of a
		50-yr high-resolution global dataset of meteorological forcings for land
		surface modeling, J. Climate, 19 (13), 3088-3111.
Variables:	Precipitation (prcp)
		Air temperature at 2m above ground (tas)
		Downward longwave at surface (dlwrf)
		Downward shortwave at surface (dswrf)
		Surface pressure (pres)
		Specific humidity (shum)
		Windspeed (wind)
Formats:        NetCDF
History:	Original daily product averaged to monthly means
Volume:         3 MB/year/variable
Date Created:   Fri Jul 14 16:12:47 EDT 2006
Change Log:     

11) Fri Jun 12 12:40:53 EDT 2009
Unrealistic downward shortwave values (dswrf) over parts of the Amazon   
and other isolated points have been corrected. All 3hourly, daily and
monthly dswrf files have been updated.

10) Mon Jun  1 10:24:49 EDT 2009
Complete update of the dataset (all files updated), including: 
i) extension to 2006; 
ii) improved sampling procedure for correction of rain day statistics; 
iii) use of latest versions of CRU (TS3.0), SRB (V3.0) and TRMM products; 
iv) improved consistency between specific and relative humidity and air temperature.

9) Mon Aug 11 16:31:01 EDT 2008
The algorithm for downscaling specific humidity was inconsistent with 
the downscaled relative humidity. This resulted in some unrealistic values 
(RH >> 100%). This has been corrected and all shum_3hourly* files have
been updated.

8) Fri Jul 25 17:12:20 EDT 2008
Daily data files for 2001-2006 (prcp, tmin, tmax, wind) have been added.
These data are experimental are are likely to change. The data
have been masked over land areas and gzip compressed. Other
variables (tas, dlwrf, dswrf, pres, shum) will be added soon. 

7) Tue Feb 26 15:31:49 EST 2008
The dswrf and dlwrf files were scaled incorrectly resulting in no
inter-annual variability. This has been corrected in all
dswrf_3hourly* and dlwrf_3hourly* files.

6) Wed Nov 14 16:20:08 EST 2007
The precipitation values in all prcp_daily*.nc files had the
wrong units and are therefore three times too small for the
units given in the metadata (mm/s). Also, the data are
the version without corrections for gauge undercatch version,
although the metadata says that they are undercatch corrected.
These two errors have been corrected and the data in ALL
prcp_daily*.nc now have the correct units and are corrected for
gauge undercatch.

5) Fri May 18 13:55:41 EDT 2007
The time variable in pres_3hourly_1960-1960.nc had some
strange values, which have been corrected:
time(1213) has value: 66878.9
time(1216) has value: 66559.87
time(1248) has value: 66340.7
time(1284) has value: 66091.77

4) Thu Nov 16 11:38:09 EST 2006
A few unrealistically low and high air temperature values occurred over
Greenland. These have been corrected and all tas_3hourly* files have been
updated.

3) Wed Nov 15 12:15:58 EST 2006
Some physically unrealistic and spurious downward shortwave values
were generated when scaling to match the SRB monthly values, especially
at high latitudes during winter. These have been corrected in all
dswrf_3hourly* files.

2) Mon Nov 13 14:08:42 EST 2006
A second version of the precipitation dataset has been added
which is not corrected for gauge undercatch. The file name
prefix is "prcp_nuc" (No Undercatch Correction).

1) Wed Nov  8 11:25:54 EST 2006
A formatting error caused missing data in the air temperature file for 1949
(tas_3hourly_1949-1949.nc, record 274). The file was corrected.