Skip to main content
Have a personal or library account? Click to login
Multigrid methods for improving the variational data assimilation in numerical weather prediction Cover

Multigrid methods for improving the variational data assimilation in numerical weather prediction

Open Access
|Dec 2014

Figures & Tables

Fig. 1

The smoothing property.

Fig. 2

Schedule of grids for (a) V-cycle, (b) W-cycle and (c) FMV-cycle.

Fig. 3

The minimisation procedure of the incremental 3D-Var.

Fig. 4

The horizontal grid structures of WRF. (a) WRF model: Arakawa C-grid staggering, (b) WRF-Var: unstaggered Arakawa A-grid.

Fig. 5

(a) Non-weighted prolongation for CCFD, (b) weighted prolongation for CCFD.

Fig. 6

(a) The observation data, (b) the contour of observation data, (c) the analysis field of 3D-Var with the conjugate gradient method and (d) the analysis field of 3D-Var with the multigrid method.

Table 1. The result of numerical experiment 1

Number of iterationCG ∇JNumber of iterationMG ∇J
030.62536518030.6253651811.1354160710.0941901520.1242786820.0003846930.0134909130.0000016440.0016098640.0000000150.0002011850.0000000060.0000265970.0000033780.0000004290.00000005100.00000001110.00000000
Fig. 7

(a) The true state of temperature, (b) the contour of true data, (c) the observation data and (d) the contour of observation data.

Fig. 8

(a–c): The analysis fields of 3D-Var with the conjugate gradient method at each iterative step 1(a), 7(b), and 13(c), (d–f): The analysis fields of 3D-Var with the multigrid method at each iterative step 1(d), 3(e) and 5(f).

Table 2. The result of numerical experiment 2

Number of iterationCG ∇JNumber of iterationMG ∇J
020.43552705020.4355270512.6409575010.0638083020.4188588220.0002616130.0687252030.0000011740.0101991340.0000000150.0017645250.0000000060.0003116970.0000525280.0000078790.00000138100.00000020110.00000003120.00000001130.00000000
Language: English
Page range: 20217 - 20217
Submitted on: Dec 5, 2012
Accepted on: May 27, 2014
Published on: Dec 1, 2014
Published by: Stockholm University Press
In partnership with: Paradigm Publishing Services

© 2014 Youn-Hee Kang, Do Young Kwak, Kyungjeen Park, published by Stockholm University Press
This work is licensed under the Creative Commons Attribution 4.0 License.