wiki:LnDevel

Version 338 (modified by lnerger, 13 hours ago) ( diff )

--

Development page

Please only use the pages that are directly linked from the start page and within the Implementation Guide

Pages linked here are under development

New for PDAF V3.1

PDAFomi_set_searchtype

PDAF_set_seedvec

PDAF_get_seedvec

PDAF_get_rndcount

PDAF_assim_flag

PDAF_generate_rndvec

PDAF_generate_rndmat (not new, but yet not documented)

PDAF3_init_forecast (clarification for PDAF_init_forecast)

PDAF3_get_fcst_info

PDAFomi_diag_rmsd (alias for PDAFomi_diag_obs_rmsd)

PDAFomi_diag_crps

PDAF_diag_diffstats

PDAFomi_diag_diffstats

PDAF_parse

PDAF3_init_parallel

PDAF_finalize

PDAF_abort

Version 3.1 - April 16, 2026

This is a feature release with performance improvements and new functionality. All changes are backward compatible. Thus, one can use PDAF 3.1 without any changes to codes existing for PDAF 3.0.

Major changes and additions

  • Performance improvements to PDAFomi_init_dim_obs_l: We once more revisited the observation search and strongly revised it adding improved search types.
    • one can now select different search types using the routine PDAFomi_set_searchtype.
    • The new default search type is in our experiments between 2 and 5 times faster than the old implementation. However, the performance improvements is case specific.
    • In tests we obtain the best performance with a search type in which PDAF first sorts the observations along a coordinate direction and subsequently performs a bisection search, were we obtain a speedup up to a factor of 30. Since the observations are reordered the assimilation result can be slightly different. Due to this we did not make this the default.
    • Recommendation: If you see in your application that the observation search in PDAFomi_init_dim_obs_l takes significant time, we recommend use PDAFomi_set_searchtype and test the different search types to find the optimal one for your application.
  • Added OMI routine PDAFomi_diag_crps: Compute the CRPS comparing the observations with the observed ensemble mean
  • Added diagnostics routine PDAF_diag_diffstats to compute statistics for comparing two vectors (correlation, centered RMSD, bias, mean absolute error, variances) analogous to PDAFomi_diag_diffstats of PDAF-OMI.
  • Added routines to control and assess random numbers seeds. The routines PDAF_set_seedvec, PDAF_get_seedvec, PDAF_get_rndcount allow one to store and reset the seed for random number generation in PDAF.
  • Added routine PDAF_generate_rndvec to perturb a vector according to a chosen distribution. This can be used, e.g. to generate perturbed model parameters.
  • Make routine PDAF_generate_rndmat accessible to user. This routine was previsouly only used internally. It is used to generate matrices representing random rotations, which, when mutiplying with it, preserve the mean state and variances of an ensemble. This can be used to generate perturbed fields, e.g., for an ensemble of perturbed forcings.
  • Added routine PDAF3_init_parallel to let PDAF configure the ensemble parallelization. It is used in the routine init_parallel_pdaf and moves the communicator splitting from this routine into PDAF, which simplifies the code. This also allows PDAF to handle the initialization and finalization of MPI which simplifies the implementation. For the offline coupling and for online coupling for models without parallelization, one does not need to call MPI_init in the user code any more.
  • New function PDAF_assim_flag as alternative to the subroutine PDAF_get_assim_flag. Using the function can simplify the code compared to using the subroutine.
  • New subroutine PDAF_parse. The command line parser is now provided as a PDAF library routine, so that it is no longer required to be a file in the user code.
  • New routine PDAF_abort to abort the parallel program (replaces the routine abort_parallel in the previous tutorial and template codes)

Improving consistency of subroutine names

The tutorial implementations and templates have been revised to use the new routines. We further included the use of an interface module mod_model_pdaf, which is intended to be the only place to directly include modules of the model. Thus, all user routines include only mod_model_pdaf instead of different model modules. This leads to a cleaner source code and allows us to use in many cases identical source files for online and offline coupling.

New for PDAF V3.0 final

PDAFomi_obs_op_extern

PDAFomi_set_name

PDAF3_set_parallel

PDAF3_init

PDAF_iau_set_state_pointer

PDAF_iau_set_ens_pointer

Note: See TracWiki for help on using the wiki.