Changes between Version 22 and Version 23 of PDAF_OMI_Overview


Ignore:
Timestamp:
Nov 25, 2020, 4:53:28 PM (3 months ago)
Author:
lnerger
Comment:

--

Legend:

Unmodified
Added
Removed
Modified
  • PDAF_OMI_Overview

    v22 v23  
    2424
    2525Compared to the classical variant to implement observations with PDAF, OMI is designed to reduce the coding effort for implementing the support for an observation type, its corresponding observation operator, and localization. In particular, only four routines need to be implemented for an observation type:
    26  - '''init_dim_obs'''
    27  - '''obs_op'''
    28  - '''init_dim_obs_l''' (only for domain-localized filters like LETKF and LESTKF)
    29  - '''localize_covar''' (only needed for the localized EnKF)
    30 Here, the first routine reads the observation from file and initializes the information describing the observations. The latter three routines merely call a generi routine provided by PDAF-OMI.
    31 Other functionality needed during the analyis step bases on variables that are initialized by these routines and is provided by PDAF-OMI.
     26 - '''init_dim_obs'''[[br]]
     27  Reads observations from a file and initialize the information describing the observations
     28 - '''obs_op'''[[br]]
     29  Calls an observation operator routine
     30 - '''init_dim_obs_l''' [[br]]
     31  Calls a generic routine to initialize local observations (only for domain-localized filters like LETKF and LESTKF)
     32 - '''localize_covar'''[[br]]
     33  Calls a generic routine to apply covariance localization (only needed for the localized EnKF)
     34
     35All functionality needed during the analyis step bases on variables that are initialized by these routines and is provided by PDAF-OMI.
    3236
    3337With OMI one implements one module for each observation type (like sea surface temperature data from some satellite sensor, or altimetry data). This guarantees that each observation type is handled indenpendently from the others. Thus different developers can implement observation types without interfering with the implementations by others.
     
    3741
    3842The main components of OMI are
    39  - '''observation modules''': One observation-specific Fortran module for each observation type
    40  - '''callback_obs_pdafomi.F90''': The observation-specific call-back routines are merely pass-through routines without own functionality. They are collected into this single file for compactness.
    41  - '''PDAF-OMI core routines''': These routines are part of the PDAF library and provide functionality for observation handling, localization, and observation operators
     43 - '''observation modules'''[[br]]
     44  One observation-specific Fortran module for each observation type
     45 - '''callback_obs_pdafomi.F90'''[[br]]
     46  The observation-specific call-back routines are merely pass-through routines without own functionality. They are collected into this single file for compactness.
     47 - '''PDAF-OMI core routines'''[[br]]
     48  These routines are part of the PDAF library and provide functionality for observation handling, localization, and observation operators
    4249
    4350Figure 1 shows the call structure for the analysis step with PDAF. For the analysis step, the core routines of PDAF (green; in particular PDAF_assimilate_X, with X being the chosen routine) calls different user-provided call-back functions. There are several routines, like those doing state localization, that are not related to observations (blue). OMI is concerned with those routines related to observations (red and purple).