SCL requires a continuous time series as input data.

The type of data and time step of data that is required depend on the stochastic data that is required. For example: to generate annual rainfall, a continuous annual rainfall time series is required as input data; to generate daily climate (rainfall, PET and temperature), a continuous daily rainfall, PET and temperature time series is required as input data.

All the models except the subdaily rainfall model in SCL require that there is no missing data in the time series.

Various time series data formats supported by SCL are detailed at www.toolkit.net.au/fileformats. It is recommended that comma delimited time series (.cdt) format is used for the annual, monthly and daily models. Currently only Bureau of Meteorology six-minute rainfall intensity (.bsm) file format is recommended for the sub-daily rainfall model.

As a general rule, the input data should have at least 30 years of historical climate time series.

• Graphical display of input time series and stochastically generated data
• Graphical display of the empirical distribution of stochastically generated data and historical data. These plots allow quick easy comparison of the distributional shape at various aggregation levels. For the daily models annual maxima curves are also produced - providing Depth-Frequency-Duration curve validation.
• Display of statistics of stochastically generated data and historical data
• Statistical summary and assessment of quality of stochastically generated data
• Easily retrievable file of stochastically generated climate data
• Easily retrievable file of statistics of generated and historical data.

• A basic understanding of stochastic climate data and the use of stochastic climate data with hydrological models to quantify uncertainty in environmental systems associated with climate variability are required.
• As a general rule, the input data should have at least 30 years of historical climate time series.
• For comparison of the statistics of the generated and historical data to be valid, the replicate length and the historical input data length must be the same. Sufficient replicates (at least 100) should also be generated to properly capture the sampling variability.
• Although stochastic hydrology is a matured science, new stochastic models are continually being developed, usually with marginal improvements on previous models. The models in SCL are selected because of available expertise, their robustness, and extensive and successful model testing using data from across Australia.