Under construction.

This combines information from:

Validation: EMOS 4.3.2
EMOS-216 issue: interpolation error analysis (Memo by Pedro Maciel)
Presentation given at December 2015 FD-RD meeting by Pedro Maciel

The information presented here will be made available to the public under the History of EMOSLIB changes.

While addressing sub-optimal performance issues for interpolating from the octahedral reduced Gaussian grid introduced with IFS cycle 41r2 to regular latitude-longitude grids, an issue was discovered with the method used to calculate the longitudinal points in the source grid. In some specific cases, this issue leads to an incorrect computation of interpolation weights and hence to incorrect interpolated values at some points of the output grid.

The issue affects all versions of EMOSLIB prior to cycle 000430. The problem is also present in fields retrieved with versions of MARS (including the WebAPI) that use and EMOSLIB prior to cycle 000430, Metview versions prior to 4.6.1 and for disseminated products from IFS cycle 41r1 and earlier.

This page provides information about the problem and the specific cases where differences can occur.

Description of the problem

The problem originates in the calculation of the longitude values of the source grid in EMOSLIB routines IRDIWE and IGDIWE. These routines use an integer value of the longitudinal grid increment (the 'stride') which, in some cases, results in a truncated value of the increment. This can result in an error in the computation of the longitude values in the source grid.

The error occurs whenever the grid spacing Dj=360/Nj at a specific latitude line, j, with Nj longitude points has a remainder smaller than 10 microdegrees (1/100000 of a degree).
The error accumulates linearly from 0° to 360° along the line of latitude.
The error is minimal at grid points to the east and maximal at those to the west of the 0° meridian.
The error is larger the greater the number of longitude points (Nj) along the line of latitude.

The incorrect computation of the longitude values leads to two potential issues:

The longitude points in the source grid are computed incorrectly. This leads to an incorrect computation of interpolation weights and hence to incorrect values at some points of the output grid.
The nearest grid points used for the interpolation may be incorrectly identified due to a numerical 'shift' of the input grid cell.

The error is most evident for parameters where the gradient of the field is large and where a change in the nearest grid points or the interpolation weights used thus has a larger effect.

Which interpolations are affected ?

The problem affects interpolations from:

input original reduced Gaussian grid point fields to output regular Gaussian or regular latitude-longitude grids (without rotation)
input octahedral reduced Gaussian grid point fields to output regular Gaussian or regular latitude-longitude grids (without rotation).

Which interpolations are not affected ?

Interpolations to rotated latitude-longitude grids are unaffected
Transformations from spherical harmonic components to grid point fields are unaffected.
Wave ?

Where is the problem fixed ?

The issue is fixed in EMOSLIB cycle 000430 and newer. This is used by:

MARS client to be updated in January 2016 (until then it is available with "mars -t")
Metview version 4.6.1 and newer
Product generation for IFS cycle 41r2 (to be implemented in Spring 2016)

Description of the problem

Which interpolations are affected ?

Which interpolations are not affected ?

Where is the problem fixed ?

Examples