NEDAS.assim_tools.assimilators

NEDAS.assim_tools.assimilators.base module

NEDAS.assim_tools.assimilators.batch module

class NEDAS.assim_tools.assimilators.batch.BatchAssimilator(c: Context)

Bases: Assimilator

init_partitions(c: Context) → list

Generate spatial partitioning of the domain partitions: dict[par_id, tuple(istart, iend, di, jstart, jend, dj)] for each partition indexed by par_id, the tuple contains indices for slicing the domain Using regular slicing is more efficient than fancy indexing (used in irregular grid)

assign_obs(c: Context) → dict

Assign the observation sequence to each partition par_id

assign_obs_to_tiles(c, state, obs, obs_rec_id)

distribute_partitions(c: Context)

Distribute partitions across processors

assimilation_algorithm(c: Context)

batch assimilation solves the matrix version EnKF analysis for each local state, the local states in each partition are processed in parallel

abstractmethod local_analysis(c, loc_id, ind, hlfactor, state_data, obs_data)

Local analysis scheme for each model state variable (grid point) to be implemented by derived classes

NEDAS.assim_tools.assimilators.ETKF module

class NEDAS.assim_tools.assimilators.ETKF.ETKFAssimilator(c: Context)

Bases: BatchAssimilator

local_analysis(c, loc_id, ind, hlfactor, state_data, obs_data)

Local analysis scheme for each model state variable (grid point) to be implemented by derived classes

NEDAS.assim_tools.assimilators.TopazDEnKF module

class NEDAS.assim_tools.assimilators.TopazDEnKF.TopazDEnKFAssimilator(c: Context)

Bases: BatchAssimilator

rfactor: float

kfactor: float

nlobs_max: int | None

local_analysis(c, loc_id, ind, hlfactor, state_data, obs_data)

Local analysis scheme for each model state variable (grid point) to be implemented by derived classes

NEDAS.assim_tools.assimilators.serial module

class NEDAS.assim_tools.assimilators.serial.SerialAssimilator(c: Context)

Bases: Assimilator

Subclass for serial assimilation algorithms

init_partitions(c: Context) → list

Generate spatial partitioning of the domain

assign_obs(c: Context)

Assign the observation sequence to each partition par_id

Parameters:

c (Context) – the runtime context object

Returns:

Indices in the full obs_seq for the subset of obs that belongs to partition par_id

Return type:

dict[ObsRecordID, dict[PartitionID, np.ndarray]]

distribute_partitions(c: Context)

Distribute partitions across processors

assimilation_algorithm(c: Context)

Implementation of the serial assimilation algorithm.

Notes

serial assimilation goes through the list of observations one by one for each obs the near by state variables are updated one by one. so each update is a scalar problem, which is solved in 2 steps: obs_increment, update_ensemble

abstractmethod obs_increment(obs_prior, obs, obs_err) → ndarray

Compute observation-space analysis increments.

Parameters:

• obs_prior (np.ndarray) – Observation priors, 1D array of type int size nens

• obs (float) – The real observation value

• obs_err (float) – Observation error std

Returns:

observation-space analysis increments

Return type:

ndarray

abstractmethod update_local_state(state_prior, obs_prior, obs_incr, state_h_dist, state_v_dist, state_t_dist, hroi, vroi, troi, h_local_func, v_local_func, t_local_func) → None

Update the local state vector with the analysis increments.

Parameters:

• state_data (np.ndarray[(nens, nfld, nloc), float]) – Local state vector

• obs_prior (np.ndarray[nens, float]) – Observation priors

• obs_incr (np.ndarray[nens, float]) – Analysis increments

abstractmethod update_local_obs(obs_data_prior, obs_used, obs_prior, obs_incr, obs_h_dist, obs_v_dist, obs_t_dist, hroi, vroi, troi, h_local_func, v_local_func, t_local_func) → None

Update the local observations with analysis increments.

Parameters:

• obs_data_prior (np.ndarray)

• (np.nd (obs_used)

NEDAS.assim_tools.assimilators.EAKF module

class NEDAS.assim_tools.assimilators.EAKF.EAKFAssimilator(c: Context)

Bases: SerialAssimilator

obs_increment(obs_prior, obs, obs_err)

Compute observation-space analysis increments.

Parameters:

• obs_prior (np.ndarray) – Observation priors, 1D array of type int size nens

• obs (float) – The real observation value

• obs_err (float) – Observation error std

Returns:

observation-space analysis increments

Return type:

ndarray

update_local_state(state_prior, obs_prior, obs_incr, state_h_dist, state_v_dist, state_t_dist, hroi, vroi, troi, h_local_func, v_local_func, t_local_func) → None

Update the local state vector with the analysis increments.

Parameters:

• state_data (np.ndarray[(nens, nfld, nloc), float]) – Local state vector

• obs_prior (np.ndarray[nens, float]) – Observation priors

• obs_incr (np.ndarray[nens, float]) – Analysis increments

update_local_obs(obs_data_prior, obs_used, obs_prior, obs_incr, obs_h_dist, obs_v_dist, obs_t_dist, hroi, vroi, troi, h_local_func, v_local_func, t_local_func) → None

Update the local observations with analysis increments.

Parameters:

• obs_data_prior (np.ndarray)

• (np.nd (obs_used)