The computational model GABRIEL is used in forecasting or producing scenarios of hydrologic inflows from different families of models adjusted to the streamflow measurements of the gauging stations. Various options can be chosen at the different phases of the time series analysis, namely: (i) model selection, (ii) parameter estimation, (iii) generation of scenarios or forecasting, and (iv) verification and validation of the selected model.

The statistical tests of GABRIEL can also be used for the verification / validation of scenarios produced by external models.

The scenarios of natural or incremental inflows can be generated by GABRIEL using annual, monthly or weekly time steps. They can be used by applications with different objectives, such as: energy planning, water resource management, among others. GABRIEL, for instance, is integrated with SDDP for the optimal operation planning of hydrothermal systems. Some features of the model GABRIEL include:

	Generation of inflow scenarios with a nonparametric method (no probability distribution);
	Set of models for selected stations and disaggregation techniques applied to others;
	Time disaggregation: annual flows can be broken down into monthly / weekly flows;
	Generalized regression: the dependent variables can be aggregated in various ways.