[NARCCAP-discuss] Bias Correction of weather data

Seth McGinnis mcginnis at ucar.edu
Fri Apr 5 16:05:21 MDT 2013


Hi Naga,

You should always assume that the model output may have a significant bias.
There are some ways of using it in an impacts model that can cause it to drop
out, but in general, yes, it's good to bias-correct the data.

What kind of bias correction you want to do will depend on how you're using
the data and on characteristics of the observational data you're using to
correct it.  Worst-case, you can always use a delta method (difference
current climatology from future climatology, apply the result to observations),
which is not very sophisticated but is widely used.

If you're interested in precipitation and hydrology, I found this paper
comparing
different methods by Teutschbein & Seiber in the Journal of Hydrology pretty 
good: http://dx.doi.org/10.1016/j.jhydrol.2012.05.052

Cheers,

--Seth


On Mon, 25 Mar 2013 14:22:34 +0000
 Naga Modala <raghuravi23 at tamu.edu> wrote:
>Dear NARCCAP staff/users,
>
>Is it always necessary to do bias correction before using the climate data
>(downloaded from NARCCAP) for modeling purposes ? Can any one suggest some
>literature review studies on bias correction of weather data ?
>
>Thanks
>
>Naga Raghuveer Modala
>Texas A&M University
>_______________________________________________
>narccap-discuss mailing list
>narccap-discuss at mailman.ucar.edu
>http://mailman.ucar.edu/mailman/listinfo/narccap-discuss



More information about the narccap-discuss mailing list