| | 1 | = Implementation of the Analysis step for the ENSRF/EAKF (Ensemble Square Root/Adjustment Filter) = |
| | 2 | |
| | 3 | {{{ |
| | 4 | #!html |
| | 5 | <div class="wiki-toc"> |
| | 6 | <h4>Implementation Guide</h4> |
| | 7 | <ol><li><a href="ImplementationGuide">Main page</a></li> |
| | 8 | <li><a href="AdaptParallelization">Adaptation of the parallelization</a></li> |
| | 9 | <li><a href="InitPdaf">Initialization of PDAF</a></li> |
| | 10 | <li><a href="ModifyModelforEnsembleIntegration">Modifications for ensemble integration</a></li> |
| | 11 | <li><a href="ImplementationofAnalysisStep">Implementation of the analysis step</a></li> |
| | 12 | <ol> |
| | 13 | <li><a href="ImplementAnalysisestkf">Implementation for ESTKF</a></li> |
| | 14 | <li><a href="ImplementAnalysislestkf">Implementation for LESTKF</a></li> |
| | 15 | <li><a href="ImplementAnalysisetkf">Implementation for ETKF</a></li> |
| | 16 | <li><a href="ImplementAnalysisletkf">Implementation for LETKF</a></li> |
| | 17 | <li><a href="ImplementAnalysisseik">Implementation for SEIK</a></li> |
| | 18 | <li><a href="ImplementAnalysislseik">Implementation for LSEIK</a></li> |
| | 19 | <li><a href="ImplementAnalysisseek">Implementation for SEEK</a></li> |
| | 20 | <li><a href="ImplementAnalysisenkf">Implementation for EnKF</a></li> |
| | 21 | <li><a href="ImplementAnalysislenkf">Implementation for LEnKF</a></li> |
| | 22 | <li>Implementation for ENSRF/EAKF</li> |
| | 23 | <li><a href="ImplementAnalysisnetf">Implementation for NETF</a></li> |
| | 24 | <li><a href="ImplementAnalysislnetf">Implementation for LNETF</a></li> |
| | 25 | <li><a href="ImplementAnalysispf">Implementation for PF</a></li> |
| | 26 | <li><a href="ImplementAnalysis_3DVar_classical">Implementation for 3D-Var</a></li> |
| | 27 | <li><a href="ImplementAnalysis_3DEnVar_classical">Implementation for 3D Ensemble Var</a></li> |
| | 28 | <li><a href="ImplementAnalysis_Hyb3DVar_classical">Implementation for Hybrid 3D-Var</a></li> |
| | 29 | </ol> |
| | 30 | <li><a href="AddingMemoryandTimingInformation">Memory and timing information</a></li> |
| | 31 | <li><a href="EnsembleGeneration">Ensemble Generation</a></li> |
| | 32 | <li><a href="DataAssimilationDiagnostics">Diagnostics</a></li> |
| | 33 | </ol> |
| | 34 | </div> |
| | 35 | }}} |
| | 36 | |
| | 37 | [[PageOutline(2-3,Contents of this page)]] |
| | 38 | |
| | 39 | || This page describes the implementation of the analysis step without using PDAF-OMI. Please see the [wiki:ImplementationofAnalysisStep page on the analysis with OMI] for the more modern and efficient implementation variant using PDAF-OMI. || |
| | 40 | |
| | 41 | The LEnKF was added with verson 3.0 of PDAF. |
| | 42 | |
| | 43 | == Overview == |
| | 44 | |
| | 45 | The ENSRF and EAKF are ensemble Kalman filter variants using serial observation processing. The implementation follows Houtekamer and Mitchell (2002) for the ENSRF and Anderson (2003) for the EAKF variant using local least squares regression. |
| | 46 | |
| | 47 | For the analysis step of the ENSRF and EAKF different operations related to the observations are needed. These operations are requested by PDAF by calling user-supplied routines. Intentionally, the operations are split into separate routines in order to keep the operations rather elementary. This procedure should simplify the implementation. The names of the required routines are specified in the call to the routine `PDAF_assimilate_ensrf` for the fully-parallel and flexible parallelization implementations (alternatively `PDAF_put_state_lenkf` for the 'flexible' implementation). With regard to the parallelization, all these routines are executed by the filter processes (`filterpe=.true.`) only. |
| | 48 | |
| | 49 | For completeness we discuss here all user-supplied routines that are specified in the interface to `PDAF_assimilate_enskf`. Thus, some of the user-supplied routines that are explained on the page explaining the modification of the model code for the ensemble integration are repeated here. |
| | 50 | |
| | 51 | In our study Nerger et al., 2015) we discussed that applying localization can lead to stability issues of the ENSRF. The filter performs a loop over all single observations and with localization the assimilation result depends on the order in which the observations are assimilated. This actually result in the effect that the assimilation result at some grid point does not only depend on the observations with the localization radius **r**, but also on observations further away if the influence the state close to the observations at distance **r** if those observations are assimilated before the observations within the radius **r**. This effect has implications on the parallelization since keeping the observation order constant does lead to a pertial serialization of the algorithm. In the implementation in PDAF we use the parallelization approach that does not guarantee the some order of the observations. Usually, the differences are small, but the benefit is a better scaling since the serialization is avoided. Nonetheless, we generally recommend using LESTKF or LETKF, or their global variants ESTKF or ETKF, since they no not depend explicitly on the observation order, and they allow for non-diagonal observation error covariance matrices. However, the ENSRF/EAKF might have a good compute performance. |
| | 52 | |
| | 53 | == `PDAF_assimilate_ensrf` == |
| | 54 | |
| | 55 | The general aspects of the filter specific routines `PDAF_assimilate_*` have been described on the page [ModifyModelforEnsembleIntegration Modification of the model code for the ensemble integration] and its sub-page on [InsertAnalysisStep inserting the analysis step]. The routine is used in the fully-parallel and the flexible implementation variant of the data assimilation system. When the offline model is used the routines `PDAF_put_state_*' are used. These have also been used in previous PDAF releases for the 'flexible' implementation variant. Here, we list once more the full interface of the routine. Subsequently, the full set of user-supplied routines specified in the call to `PDAF_assimilate_ensrf` is explained. |
| | 56 | |
| | 57 | The interface when using the LEnKF is the following: |
| | 58 | {{{ |
| | 59 | SUBROUTINE PDAF_assimilate_lenkf(U_collect_state, U_distribute_state, U_init_dim_obs, & |
| | 60 | U_obs_op, U_init_obs, U_prepoststep, U_localize, & |
| | 61 | U_add_obs_err, U_init_obscovar, U_next_observation, status) |
| | 62 | }}} |
| | 63 | with the following arguments: |
| | 64 | * [#U_collect_statecollect_state_pdaf.F90 U_collect_state]: The name of the user-supplied routine that initializes a state vector from the array holding the ensemble of model states from the model fields. This is basically the inverse operation to `U_distribute_state` used in `PDAF_get_state` as well as here. |
| | 65 | * [#U_distribute_statedistribute_state_pdaf.F90 U_distribute_state]: The name of a user supplied routine that initializes the model fields from the array holding the ensemble of model state vectors. |
| | 66 | * [#U_init_dim_obsinit_dim_obs_pdaf.F90 U_init_dim_obs]: The name of the user-supplied routine that provides the size of observation vector |
| | 67 | * [#U_obs_opobs_op_pdaf.F90 U_obs_op]: The name of the user-supplied routine that acts as the observation operator on some state vector |
| | 68 | * [#U_init_obsinit_obs_pdaf.F90 U_init_obs]: The name of the user-supplied routine that initializes the vector of observations |
| | 69 | * [#U_prepoststepprepoststep_ens_pdaf.F90 U_prepoststep]: The name of the pre/poststep routine as in `PDAF_get_state` |
| | 70 | * [#U_localizelocalize_covar_pdaf.F90 U_localize]: Apply covariance localization to the matrices HP and HPH^T^ |
| | 71 | * [#U_add_obs_erradd_obs_err_pdaf.F90 U_add_obs_err]: The name of the user-supplied routine that adds the observation error covariance matrix to the ensemble covariance matrix projected onto the observation space. |
| | 72 | * [#U_init_obscovarinit_obscovar_pdaf.F90 U_init_obscovar]: The name of the user-supplied routine that initializes the observation error covariance matrix. |
| | 73 | * [#U_next_observationnext_observation_pdaf.F90 U_next_observation]: The name of a user supplied routine that initializes the variables `nsteps`, `timenow`, and `doexit`. The same routine is also used in `PDAF_get_state`. |
| | 74 | * `status`: The integer status flag. It is zero, if `PDAF_assimilate_lenkf` is exited without errors. |
| | 75 | |
| | 76 | == `PDAF_put_state_lenkf` == |
| | 77 | |
| | 78 | When the 'flexible' implementation variant is chosen for the assimilation system, the routine `PDAF_put_state_lenkf` has to be used instead of `PDAF_assimilate_lenkf`. The general aspects of the filter specific routines `PDAF_put_state_*` have been described on the page [ModifyModelforEnsembleIntegration Modification of the model code for the ensemble integration]. The interface of the routine is identical with that of `PDAF_assimilate_lenkf` with the exception the specification of the user-supplied routines `U_distribute_state` and `U_next_observation` are missing. |
| | 79 | |
| | 80 | The interface when using the LEnKF is the following: |
| | 81 | {{{ |
| | 82 | SUBROUTINE PDAF_put_state_lenkf(U_collect_state, U_init_dim_obs, U_obs_op, & |
| | 83 | U_init_obs, U_prepoststep, U_localize, & |
| | 84 | U_add_obs_err, U_init_obscovar, status) |
| | 85 | }}} |
| | 86 | |
| | 87 | |
| | 88 | == User-supplied routines == |
| | 89 | |
| | 90 | Here all user-supplied routines are described that are required in the call to `PDAF_assimilate_lenkf`. For some of the generic routines, we link to the page on [ModifyModelforEnsembleIntegration modifying the model code for the ensemble integration]. |
| | 91 | |
| | 92 | To indicate user-supplied routines we use the prefix `U_`. In the tutorials in `tutorial/` and in the template directory `templates/` these routines exist without the prefix, but with the extension `_pdaf`. The files are named correspondingly. In the section titles below we provide the name of the template file in parentheses. |
| | 93 | |
| | 94 | In the subroutine interfaces some variables appear with the suffix `_p`. This suffix indicates that the variable is particular to a model sub-domain, if a domain decomposed model is used. Thus, the value(s) in the variable will be different for different model sub-domains. |
| | 95 | |
| | 96 | |
| | 97 | === `U_collect_state` (collect_state_pdaf.F90) === |
| | 98 | |
| | 99 | This routine is independent of the filter algorithm used. |
| | 100 | See the page on [InsertAnalysisStep#U_collect_statecollect_state_pdaf.F90 inserting the analysis step] for the description of this routine. |
| | 101 | |
| | 102 | |
| | 103 | === `U_distribute_state` (distribute_state_pdaf.F90) === |
| | 104 | |
| | 105 | This routine is independent of the filter algorithm used. |
| | 106 | See the page on [InsertAnalysisStep#U_distribute_statedistribute_state_pdaf.F90 inserting the analysis step] for the description of this routine. |
| | 107 | |
| | 108 | |
| | 109 | === `U_init_dim_obs` (init_dim_obs_pdaf.F90) === |
| | 110 | |
| | 111 | This routine is used by all global filter algorithms (SEEK, SEIK, EnKF, ETKF) and by the LEnKF. |
| | 112 | |
| | 113 | The interface for this routine is: |
| | 114 | {{{ |
| | 115 | SUBROUTINE init_dim_obs(step, dim_obs_p) |
| | 116 | |
| | 117 | INTEGER, INTENT(in) :: step ! Current time step |
| | 118 | INTEGER, INTENT(out) :: dim_obs_p ! Dimension of observation vector |
| | 119 | }}} |
| | 120 | |
| | 121 | The routine is called at the beginning of each analysis step. It has to initialize the size `dim_obs_p` of the observation vector according to the current time step. Without parallelization `dim_obs_p` will be the size for the full model domain. When a domain-decomposed model is used, `dim_obs_p` will be the size of the observation vector for the sub-domain of the calling process. |
| | 122 | |
| | 123 | Some hints: |
| | 124 | * It can be useful to not only determine the size of the observation vector at this point. One can also already gather information about the locations of the observations, which will be used later, e.g. to implement the observation operator. An array for the locations can be defined in a module like `mod_assimilation` of the example implementation. |
| | 125 | |
| | 126 | |
| | 127 | === `U_obs_op` (obs_op_pdaf.F90) === |
| | 128 | |
| | 129 | This routine is used by all global filter algorithms (SEEK, SEIK, EnKF, ETKF) and the LEnKF. |
| | 130 | |
| | 131 | The interface for this routine is: |
| | 132 | {{{ |
| | 133 | SUBROUTINE obs_op(step, dim_p, dim_obs_p, state_p, m_state_p) |
| | 134 | |
| | 135 | INTEGER, INTENT(in) :: step ! Currrent time step |
| | 136 | INTEGER, INTENT(in) :: dim_p ! PE-local dimension of state |
| | 137 | INTEGER, INTENT(in) :: dim_obs_p ! Dimension of observed state |
| | 138 | REAL, INTENT(in) :: state_p(dim_p) ! PE-local model state |
| | 139 | REAL, INTENT(out) :: m_state_p(dim_obs_p) ! PE-local observed state |
| | 140 | }}} |
| | 141 | |
| | 142 | The routine is called during the analysis step. It has to perform the operation of the observation operator acting on a state vector that is provided as `state_p`. The observed state has to be returned in `m_state_p`. |
| | 143 | |
| | 144 | For a model using domain decomposition, the operation is on the PE-local sub-domain of the model and has to provide the observed sub-state for the PE-local domain. |
| | 145 | |
| | 146 | Hint: |
| | 147 | * If the observation operator involves a global operation, e.g. some global integration, while using domain-decomposition one has to gather the information from the other model domains using MPI communication. |
| | 148 | |
| | 149 | |
| | 150 | === `U_init_obs` (init_obs_pdaf.F90) === |
| | 151 | |
| | 152 | This routine is used by all global filter algorithms (SEEK, SEIK, EnKF, ETKF) and the LEnKF. |
| | 153 | |
| | 154 | The interface for this routine is: |
| | 155 | {{{ |
| | 156 | SUBROUTINE init_obs(step, dim_obs_p, observation_p) |
| | 157 | |
| | 158 | INTEGER, INTENT(in) :: step ! Current time step |
| | 159 | INTEGER, INTENT(in) :: dim_obs_p ! PE-local dimension of obs. vector |
| | 160 | REAL, INTENT(out) :: observation_p(dim_obs_p) ! PE-local observation vector |
| | 161 | }}} |
| | 162 | |
| | 163 | The routine is called during the analysis step. |
| | 164 | It has to provide the vector of observations in `observation_p` for the current time step. |
| | 165 | |
| | 166 | For a model using domain decomposition, the vector of observations that exist on the model sub-domain for the calling process has to be initialized. |
| | 167 | |
| | 168 | |
| | 169 | === `U_prepoststep` (prepoststep_ens_pdaf.F90) === |
| | 170 | |
| | 171 | The general aspects of this routines have already been described on the [ModifyModelforEnsembleIntegration#U_prepoststepprepoststep_ens_pdaf.F90 page on modifying the model code for the ensemble integration] for the SEIK filter. For completeness, the description is repeated specifically for the EnKF: |
| | 172 | |
| | 173 | The interface of the routine is identical for all filters, but sizes can vary. Also, the particular operations that are performed in the routine can be specific for each filter algorithm. |
| | 174 | |
| | 175 | The interface for this routine is for the LEnKF |
| | 176 | {{{ |
| | 177 | SUBROUTINE prepoststep(step, dim_p, dim_ens, dim_ens_p, dim_obs_p, & |
| | 178 | state_p, Uinv, ens_p, flag) |
| | 179 | |
| | 180 | INTEGER, INTENT(in) :: step ! Current time step |
| | 181 | ! (When the routine is called before the analysis -step is provided.) |
| | 182 | INTEGER, INTENT(in) :: dim_p ! PE-local state dimension |
| | 183 | INTEGER, INTENT(in) :: dim_ens ! Size of state ensemble |
| | 184 | INTEGER, INTENT(in) :: dim_ens_p ! PE-local size of ensemble |
| | 185 | INTEGER, INTENT(in) :: dim_obs_p ! PE-local dimension of observation vector |
| | 186 | REAL, INTENT(inout) :: state_p(dim_p) ! PE-local forecast/analysis state |
| | 187 | ! The array 'state_p' is not generally not initialized in the case of SEIK/EnKF/ETKF. |
| | 188 | ! It can be used freely in this routine. |
| | 189 | REAL, INTENT(inout) :: Uinv(1, 1) ! Not used not LEnKF |
| | 190 | REAL, INTENT(inout) :: ens_p(dim_p, dim_ens) ! PE-local state ensemble |
| | 191 | INTEGER, INTENT(in) :: flag ! PDAF status flag |
| | 192 | }}} |
| | 193 | |
| | 194 | The routine `U_prepoststep` is called once at the beginning of the assimilation process. In addition, it is called during the assimilation cycles before the analysis step and after the ensemble transformation. The routine is called by all filter processes (that is `filterpe=1`). |
| | 195 | |
| | 196 | The routine provides for the user the full access to the ensemble of model states. Thus, user-controlled pre- and post-step operations can be performed. For example the forecast and the analysis states and ensemble covariance matrix can be analyzed, e.g. by computing the estimated variances. In addition, the estimates can be written to disk. |
| | 197 | |
| | 198 | Hint: |
| | 199 | * If a user considers to perform adjustments to the estimates (e.g. for balances), this routine is the right place for it. |
| | 200 | * Only for the SEEK filter the state vector (`state_p`) is initialized. For all other filters, the array is allocated, but it can be used freely during the execution of `U_prepoststep`. |
| | 201 | * The array `Uinv` is not used in the EnKF. Internally to PDAF, it is allocated to be of size (1,1). |
| | 202 | * Apart from the size of the array `Uinv`, the interface is identical for all ensemble filters (SEIK/ETKF/EnKF/LSEIK/LETKF/LEnKF). In general it should be possible to use an identical pre/poststep routine for all these filters. |
| | 203 | * The interface through which `U_prepoststep` is called does not include the array of smoothed ensembles. In order to access the smoother ensemble array one has to set a pointer to it using a call to the routine `PDAF_get_smootherens` (see page on [AuxiliaryRoutines auxiliary routines]) |
| | 204 | |
| | 205 | |
| | 206 | |
| | 207 | === `U_localize` (localize_covar_pdaf.F90) === |
| | 208 | |
| | 209 | This routine is only used for the LEnKF. |
| | 210 | |
| | 211 | The interface for this routine is: |
| | 212 | {{{ |
| | 213 | SUBROUTINE U_localize(dim_p, dim_obs, HP, HPH) |
| | 214 | |
| | 215 | INTEGER, INTENT(in) :: dim_p ! PE-local state dimension |
| | 216 | INTEGER, INTENT(in) :: dim_obs ! Dimension of global observation vector |
| | 217 | REAL, INTENT(inout) :: HP(dim_obs, dim_p) ! Matrix HP |
| | 218 | REAL, INTENT(inout) :: HPH(dim_obs, dim_obs) ! Matrix HPH^T^ |
| | 219 | }}} |
| | 220 | |
| | 221 | The routine is called during the analysis step and has to apply the element-wise Schur product for the covariance localization of the two matrices'''HP''' and '''HPH^T^''', which are provided as input/output arguments. |
| | 222 | |
| | 223 | Notes: |
| | 224 | * In case of a parallelization with domain decomposition, `HP` contains only the columns of the matrix that resides on the model sub-domain of the calling process. The number of rows is that of the global number of observations |
| | 225 | |
| | 226 | Hints: |
| | 227 | * To compute the localization one can use the routine `PDAF_local_weight` after computing the distance between two elements in the matrix '''HP''' or '''HPH^T^'''. |
| | 228 | |
| | 229 | |
| | 230 | === `U_add_obs_err` (add_obs_err_pdaf.F90) === |
| | 231 | |
| | 232 | This routine is only used for the EnKF and LEnKF. |
| | 233 | |
| | 234 | The interface for this routine is: |
| | 235 | {{{ |
| | 236 | SUBROUTINE add_obs_err(step, dim_obs, C) |
| | 237 | |
| | 238 | INTEGER, INTENT(in) :: step ! Current time step |
| | 239 | INTEGER, INTENT(in) :: dim_obs ! Dimension of obs. vector |
| | 240 | REAL, INTENT(inout) :: C(dim_obs, dim_obs) ! Matrix to that the observation |
| | 241 | ! error covariance matrix is added |
| | 242 | }}} |
| | 243 | |
| | 244 | The routine is called during the analysis step. During the analysis step of the LEnKF, the projection of the ensemble covariance onto the observation space is computed. This matrix is provided to the routine as `C_p`. The routine has to add the observation error covariance matrix to `C_p`. |
| | 245 | |
| | 246 | The operation is for the global observation space. Thus, it is independent of whether the filter is executed with or without parallelization. |
| | 247 | |
| | 248 | Hints: |
| | 249 | * The routine does not require that the observation error covariance matrix is added as a full matrix. If the matrix is diagonal, only the diagonal elements have to be added. |
| | 250 | |
| | 251 | |
| | 252 | |
| | 253 | === `U_init_obscovar` (init_obscovar_pdaf.F90) === |
| | 254 | |
| | 255 | This routine is only used for the EnKF and LEnKF. |
| | 256 | |
| | 257 | The interface for this routine is: |
| | 258 | {{{ |
| | 259 | SUBROUTINE init_obscovar(step, dim_obs, dim_obs_p, covar, m_state_p, & |
| | 260 | isdiag) |
| | 261 | |
| | 262 | INTEGER, INTENT(in) :: step ! Current time step |
| | 263 | INTEGER, INTENT(in) :: dim_obs ! Dimension of observation vector |
| | 264 | INTEGER, INTENT(in) :: dim_obs_p ! PE-local dimension of observation vector |
| | 265 | REAL, INTENT(out) :: covar(dim_obs, dim_obs) ! Observation error covariance matrix |
| | 266 | REAL, INTENT(in) :: m_state_p(dim_obs_p) ! PE-local observation vector |
| | 267 | LOGICAL, INTENT(out) :: isdiag ! Whether the observation error covar. matrix is diagonal |
| | 268 | }}} |
| | 269 | |
| | 270 | The routine is called during the analysis step and is required for the generation of an ensemble of observations. It has to initialize the global observation error covariance matrix `covar`. In addition, the flag `isdiag` has to be initialized to provide the information, whether the observation error covariance matrix is diagonal. |
| | 271 | |
| | 272 | The operation is for the global observation space. Thus, it is independent of whether the filter is executed with or without parallelization. |
| | 273 | |
| | 274 | Hints: |
| | 275 | * The local observation vector `m_state_p` is provided to the routine for the case that the observation errors are relative to the value of the observation. |
| | 276 | |
| | 277 | |
| | 278 | === `U_next_observation` (next_observation_pdaf.F90) === |
| | 279 | |
| | 280 | This routine is independent of the filter algorithm used. |
| | 281 | See the page on [InsertAnalysisStep#U_next_observationnext_observation_pdaf.F90 inserting the analysis step] for the description of this routine. |
| | 282 | |
| | 283 | |
| | 284 | |
| | 285 | == Execution order of user-supplied routines == |
| | 286 | |
| | 287 | For the :EnKF, the user-supplied routines are essentially executed in the order they are listed in the interface to `PDAF_assimilate_lenkf`. The order can be important as some routines can perform preparatory work for later routines. For example, `U_init_dim_obs` can prepare an index array that provides the information for executing the observation operator in `U_obs_op`. |
| | 288 | |
| | 289 | Before the analysis step is called the following routine is executed: |
| | 290 | 1. [#U_collect_statecollect_state_pdaf.F90 U_collect_state] |
| | 291 | |
| | 292 | The analysis step is executed when the ensemble integration of the forecast is completed. During the analysis step the following routines are executed in the given order: |
| | 293 | 1. [#U_prepoststepprepoststep_ens_pdaf.F90 U_prepoststep] (Call to act on the forecast ensemble, called with negative value of the time step) |
| | 294 | 1. [#U_init_dim_obsinit_dim_obs_pdaf.F90 U_init_dim_obs] |
| | 295 | 1. [#U_obs_opobs_op_pdaf.F90 U_obs_op] (`dim_ens` calls: one call for each ensemble member) |
| | 296 | 1. [#U_localizelocalize_covar_pdaf.F90 U_localize] |
| | 297 | 1. [#U_add_obs_erradd_obs_err_pdaf.F90 U_add_obs_err] |
| | 298 | 1. [#U_init_obsinit_obs_pdaf.F90 U_init_obs] |
| | 299 | 1. [#U_init_obscovarinit_obscovar_pdaf.F90 U_init_obscovar] |
| | 300 | 1. [#U_obs_opobs_op_pdaf.F90 U_obs_op] (`dim_ens` calls: one call for each ensemble member, repeated to reduce storage) |
| | 301 | 1. [#U_prepoststepprepoststep_ens_pdaf.F90 U_prepoststep] (Call to act on the analysis ensemble, called with (positive) value of the time step) |
| | 302 | |
| | 303 | In case of the routine `PDAF_assimilate_enkf`, the following routines are executed after the analysis step: |
| | 304 | 1. [#U_distribute_statedistribute_state_pdaf.F90 U_distribute_state] |
| | 305 | 1. [#U_next_observationnext_observation_pdaf.F90 U_next_observation] |
