

Optimal Interpolation Majid Azadi Iran Meteorological Organization Nov. 4, 2004

Abstract: 
The process of interpolating the meteorological variables from
irregular points of observed data into regular grid points is
known as objective analysis. One theoretically powerful method
described here is the optimal interpolation. The formulation
presented here is mainly bases on theoretical considerations and
not practical implementation. The purpose is to travel the
relationship between optimum interpolation and variational
schemes. The method is based on minimizing the analysis error
variance. This is done by proper choice of determining the
coefficient matrix $W$ in the following formula: $X_a=X_b+Wd$
Where $X_a$ and $X_b$ are analysis and background vectors both on
the grid points, $d$ is observational increments and $W$ is weight
or gain matrix. In principal it is possible to write the analysis
error in terms of background and observation errors. The main
difficulty however is that these errors are never known and only
some assumption can be made about their statistical properties.

Information:

Date:  Thursday, Nov. 4, 2004 at 10:00 
Place:  Lecture Hall, Niavaran Bldg., Niavaran Sqr., Tehran, Iran 

 