Select one of the options Specify black box model configuration directly in the blackBoxModelConfig format Specify filename of the black box model configuration file. File format: blackBoxModelConfig Identity of model's exchange item Specify one or more model exchange items on which to impose noise defined by this noise model's exchange item (for a single model ExchangeItem, use modelExchangeItemId in the attributes instead of this list) Identity of this noise model's exchange item Identity of model's exchange item on which to impose this noise model's exchange item Deprecated, use attribute "transformation" instead. Type of operation to performed when noise is added to an (parameter or state) exchange item: 'add' the noise to the exchange item's values, 'multiply' the exchange item's values with the noise, or 'set' the values (used in those cases where the noise in fact is a realization). Transformation of noise variable. 'identity' means that the value will be added directly without any transformation to the adjusted parameter. 'ln' means that the value is transformed logarithmically before it is added to the parameter; this gives a correction, which is a fraction of the parameter value. When adding noise to a time dependent exchange item for the next (e.g. EnKF) time slice, skip the first time step (because is was already modified in the previous time slice) When adding noise to a state exchange time, add the difference between the current noise and the previous noise, instead of adding the full noise. Integer ratio between the size of the model state and the noise model state (1 noise model value will be applied to "stateSizeNoiseSizeRatio" model values) When true, state noise will be added after compute instead of before The list of relations Specify name of xml file containing configuration of the noise model. File format depends on class. Specify list of input arguments for the noise model The relations between the noise model's exchange items and the model's exchange items The class name implementing the noise model The noise model's working directory Identity of model's exchange item Identity of noise model's exchange item Transformation of noise variable. 'identity' means that the value will be added directly without any transformation to the adjusted parameter. 'ln' means that the value is transformed logarithmically before it is added to the parameter; this gives a correction, which is a fraction of the parameter value. The standard deviation value The standard deviation value Set the initial value of this parameter (optional) Transformation of correction factor. 'identity' means that the correction will be added directly without any transformation to the adjusted parameter. 'ln' means that the correction is transformed logarithmically before it is added to the parameter; this gives a correction, which is a fraction of the parameter value. 'set' will set the correction directly as the parameter value Set the standard deviation of this parameter and the method how the parameter should be corrected The uncertainy of this parameter is specified in the uncertaintyModule mentioned in the top of the present stochastic model configuration Select the parameter vector(s) and/or subvector(s) to adjust Select the whole parameter vector Select only a part of the parameter vector Set the scaling factor for this parameter (optional) Optional id for composed items that are all modified by the same parameter. If omitted, the id's of the separate item will be concatenated. Note: the (concatenated) id is mainly used by the sparse dud algorithm Specify standard deviation for dx and dy Set the standard deviation of this parameter and the method how the parameter should be corrected Select the sets of parameter vectors to adjust, in pairs of two: amplitude and phase Select whole parameter vector Select only a part of the parameter vector Scaling factor for dx Scaling factor for dy Name suffix for dx in the noise model exchange items Name suffix for dy in the noise model exchange items Name for dx in the noise model exchange items Name for dy in the noise model exchange items Identity of the uncertain item as defined in BB model configuration file. Operation type: add (default), multiply or set ARMA constants, which reflects temporal correlation per timestep of the ARMA model. Identity of the uncertain item as defined in BB model configuration file. Operation type: add (default), multiply or set ARMA constants, which reflects temporal correlation per timestep of the ARMA model. Operation type: add (default) or multiply The standard deviation of the coloured noise to be generated by the model. Decorrelation time. Operation type: add (default) or multiply Select the way the subvector is derived from the source vector. Derive subvector by specifying indices of source vector. Derive subvector by using a configuration file or string. Set the identity of this subVector. Specify the identity of the source vector as defined in blackBoxModelConfig. Specify name of xml file containing configuration of the noise model. File format depends on class. The relations between the noise model's exchange items and the model's exchange items The model's vector that the noise will be added to. The model's vector that the noise will be added to. The noise model is fully specified in the uncertainty module The PDF of the noise model is specified in the uncertainty module, the ARMA constants are specified here Both the noise model's PDF (normal distribution) and the ARMA constants are specified here The standard deviation and the decoralation time scale of the AR1 model is specified Specify the identity of the noise model vector. It is required if the noise model contains more than one source vectors. If the noise model contains one subvector, it may be omitted. In that case the source (sub-)vector's id is used. The class that implements the noise model The working directory where the noise model is started Identity of the vector Identity of the source vector as defined in blackBoxModelConfig Select the way the subvector is derived from the source vector Select subvector using indices of the vector elements Define operator for deriving the subvector Identity of the subVector Identity of the source vector as defined in blackBoxModelConfig The uncertainty module that defines and generates the noise correction factor exchange items Specify which model parameters to adjust and how Specify which model parameters to adjust and how, under a transformation of the vector representation, from Cartesian (dx,dy) to Polar (amplitude,phase) Add a noise model to the state Add a vector to the state Add a subvector to the state Add a vector to the predictor Add a subvector to the predictor A flag for indicating whether predictor timeseries should be kept in the memory during simulation. Specify the model parameters vectors Specify the model state vectors Specify the predictor vectors. i.e. the model variables that correspond to the observer variables. This can contain three different types of mappings between model and observer variables: 1. a vector for a mapping between an observed scalar and a scalar in the model. In this case the coordinates are not used and no interpolation is done. 2. a vector for a mapping between an observed grid and a grid in the model. In this case the model values are interpolated to get values at the cell centers of the observed grid. This uses the coordinates of both grids and bilinear interpolation. 3. a subVector for a mapping between an observed scalar and a cell of a grid in the model. In this case the coordinates are not used and no interpolation is done. Specify the model directory that will contain a set of restart files for a certain time stamp (written when the algorithm requests to store states, read when the algorithm asks to restore a state) Specify the file name prefix for storing the state of the noise model(s) Specify the file name prefix for storing the model state (must end with .zip) Contains the elements in fixed order of occurrence Specify either a black box model or a black box model factory Specify a specific black box model configuration. This can be either by writing directly the model configuration or by referring to a black box model configuration file. Define a model factory. Specify class or executable, arguments to supply and fail/success checks to perform DEPRECATED. Use of this element is discouraged, use vectorSpecification -> parameters -> uncertaintyModule instead. Define stochastic model through an uncertainty module. Specify class or executable file, arguments to supply and fail/success checks to perform. Optional. Path and name of a range validation config file (relative to this configuration file) that contains range validation constraints for exchange items. Specify external forcings, being either boundaries, noise on boundaries or boundaries including noise. Define stochastic model by specifying the vectors of parameters, state, and/or predictor Restart dir./files information The data object containing the forcing data. Specify the operation when imposing the forcing ExchangeItems on the model ExchangeItems. The openda class name of the forcing's data object. Name of the forcing's data file. Input argument(s) for the corresponding openda:class (set): Replace model ExchangeItem's content by forcing ExchangeItem's content. (add): Add the forcing ExchangeItem's content to the model ExchangeItem's content. (multiply): Multiply the model ExchangeItem's content by the forcing ExchangeItem's content. List the forcing ExchangeItems for which to apply this mapping's operation. If this list is not specified, then the specified operation will be used for all forcing ExchangeItems from the specified file (in that case the model ExchangeItems must have the same ids as the forcing ExchangeItems). Identity of a forcing ExchangeItem. Identity of the model ExchangeItem on which to impose this forcing ExchangeItem. If the model ExchangeItem id equals the id of the forcing ExchangeItem, then this attribute is not needed. Type of operation to performed when noise is added to an (parameter or state) exchange item: add the noise to the exchange item's values, multiply the exchange item's values with the noise, or set the values (used in those cases where the noise in fact is a realization) Tranformation of correction factor. Two values are possible: 'identity' and 'ln'. The 'identity' means that the correction will be added directly without any transformation to the adjusted parameter. On the other hand, 'ln' means that correction is transformed logarithmically before it is added to the parameter; this gives a correction, which is a fraction of the parameter value. Configuration file for the Stochastic model class org.openda.blackbox.wrapper.BBStochModelFactory. Specifies either the deterministic model in the modelConfig element, or the model instances generator in the modelFactory element, and the stochastic model that acts as umbrella over the deterministic model, in the vectorSpecification element, and an optional restartInfo element for non-default handling of deterministic model restarts. The element uncertaintyModule is included for backwards compatibility only.