Long term and mid-term reservoir operation involves derivation of rule curves for optimal management
of the available resource. The present work deals with reservoir operation in the Aurès arid
region. As an example, Babar reservoir is selected to apply the proposed approach which estimates all
the water balance terms, especially those which are random as water inflows. For each demand scenario
a reservoir operation optimization model using Explicit Stochastic Dynamic Programming (ESDP)
is performed, to derive optimal rule curves based on historical operating records (Jan 2002–Dec 2013)
and using “Reservoir” R package®. Subsequently, risk analysis is conducted for these different demand
scenarios rules by the RRV (reliability, resilience, vulnerability) metrics. Results show the advantage
of using the “Reservoir” R package for a rapid and an easy analysis of the performance criteria
jointly with the optimization algorithm to Re-operate Reservoir operation.

Reservoir silting is one of the principal problems affecting the performance of dams in Algeria from the standpoint of reservoir capacity for storage. Foum El Kherza Reservoir (also known as Foum El Gherza), near Biskra Town, Algeria, is subject to dredging operations with the intent of recovering 70% of its initial storage capacity of 47 hm³ (million cubic metres). The forecasting of sediment volume trapped in the reservoir is essential to plan the future use of this resource and to sustain irrigation for the palm groves characteristic of the region. However, there are currently no sediment data, nor a sediment rating curve, for predicting sediment inflow based on hydrological data. This paper describes the optimization of a cumulative trapped sediment curve for Foum El Kherza Reservoir based on 44 years of daily inflows, by using a spreadsheet optimization tool, Microsoft Excel® Solver to calibrate the cumulative sediment load against the cumulative sediment inflow as documented by eight bathymetric surveys since the dam construction.