March 17, 2015:
The 3-hourly VIC simulations are complete and posted here, based on the v.1.2 
forcings, and VIC versions 4.1.2.c. A description of the file contents is 
contained in the readme.txt file in this directory.

December 26, 2014:
The daily VIC simulations are complete and posted here, based on the v.1.2 
forcings, and VIC versions 4.1.2.c. A description of the file contents is 
contained in the readme.txt file in this directory.

March 19, 2014:
The forcing data set is publicly available, however, we request that users
use the citation below when using these data. Following a series of QA/QC,
versions 1.2 has been posted (see the readme.txt file for formatting details).
Work is currently underway to run the VIC model using forcing version 1.2,
which will be posted in the coming weeks.

Version history**:

1.2: March, 2014

Resolved issues:
1. Repeated station entries in CA and TX were removed, which impacted the grid
weighting in previous versions;
2. Stations with valid Tmax *or* Tmin but with 'nodata' for the other quanitity
e.g. Tmax valid, Tmin nodata, were subsequently set to nodata to prevent
erroneous diurnal temperature ranges. 
3. Tmax, Tmin for USHCN stations were erroneously bound >= 0F (~-18C) in v.1.1;
4. A precipitation panel [-75 > lat > -80; lon > 37] was discovered in v.1.1
that suffered from precipitation issues present in v.1.0

Known issues:
1. Tmax, Tmin for a station in northeast Alabama contain large temporal 
discontinuities.
2. Number of wet-days for northern British Columbia portion of domain, as well
as northwest New York state appear very high.

1.1: January, 2014

Resolved issues:
1. Precipitation data prior to 1950 were erroneously large due to days of 
zero precipitation being set to nodata as part of the DSI-3206 data set.
This caused an overabundance of precipitaiton since the gridding code only
considers valid days.
These were corrected to zero using USHCN data, furthermore months with more
than 50% missing data for precipitation were set to zero;
2. Both Tmax and Tmin were set to nodata for station days reporting Tmax<=Tmin
3. Forcings over Canada, there was a bug in how the coordinate system,
i.e. latitude and longitude, was being read by the gridding code over Canada,
which has been resolved. This bug fix will improve both temperature and precipitation, with the former having larger improvements due to the lack of 
PRISM scaling.
4. Finally, those stations that were coincedent with USHCN stations were replaced with the QC'd values from HCN, which were identical in most cases, this 
represents ~1200 total stations. (http://cdiac.ornl.gov/epubs/ndp/ushcn/ushcn.html)

1.0: September, 2013

Resolved issues:
1. The reanalysis wind data were re-gridded by Dr. Bart Nijssen (UW), after
we found a shift of approximately 1.8 degrees north and west in the previous 
data set.  Bart's netCDF-based regridding has fixed this.  

Notes: Official release, corresponds to publication (citation below).

0.9: April, 2013

Notes: pre-release, contains erroneous 1/8 degree pattern in precipitation.

*0.8: February, 2013
* Note: versions 0.8 and 0.9 are equivalent

** Users are encouraged to use only the most up-to-date data

Livneh, B., E. A. Rosenberg, C. Lin, B. Nijssen, V. Mishra, K. M. Andreadis, E.
P. Maurer, and D. P. Lettenmaier, 2013: A Long-Term Hydrologically Based Dataset of Land Surface Fluxes and States for the Conterminous United States: Update and Extensions. J. Climate, 26, 9384~9392.