Dear Husain,

The difference between a multi-model ensemble and a grand ensemble might depend on the definitions you use. A multi-model ensemble might consist of the deterministic (or control) forecasts of each numerical weather model in the database, while a grand ensemble generally include all ensemble members of each model in the database.

It is difficult to estimate how much time it takes to prepare and downscale the seasonal forecasts for hydrological applications. Some factors influencing this estimation are:

- number of variables to include (only precipitation or also variables relevant for evapotranspiration);

- number of models and number of ensemble members for each model to include;

- geographical extent;

- downscaling procedure (e.g. statistical or dynamical);

- bias-correction technique, etc.

Cloke and Pappenberger (2009, Journal of Hydrology, 275, 613-626) give a good overview of ensemble forecasting including a discussion of these issues.

Regards,

Martijn