Skip to content

TSTool / Ensemble Processing

This documentation provides information about time series ensemble processing.


Introduction

TSTool can be used to automate processing ensembles of time series. Time series in an ensemble are represented as individual time series with the same time series identifier, but having a different sequence identifier, for example the historical year representing a simulation input. The ensemble also has a unique identifier, which is used in commands.

Example use cases include:

  • convert a historical time series into overlapping "traces" that form an ensemble
  • process an ensemble of model runs, for example representing different climate or hydrology scenarios
  • create a visualization of exceedence probabilities based on an ensemble

Commands

Ensemble processing commands are found in the Commands / Ensemble Processing menu.

Many time series processing commands allow an ensemble of time series to be processed by selecting the TSList=EnsembleID and EnsembleID command parameters.

Some commands such as ReadRiverWare support reading ensembles from input formats that can represent ensembles.

The CreateEnsembleFromOneTimeSeries command can be used to convert a historical time series into an ensemble for analysis and visualization, for example to use with other commands that can process ensembles.

TSTool User Interface

The Results / Ensembles tab lists time series ensembles that are created by running commands. Right-click on an ensemble to graph.

Examples

See the Examples section of command documentation.