The optimal operation of reservoirs is regarded as one of the most vital issues in water resources management, which also plays an important role in environmental processes and control of ecosystem pollution, especially in river water. The present study seeks to develop a new algorithm on single and multi-reservoir water resources systems, namely Atom Search Optimization (ASO). For this purpose, the algorithm is analyzed on mathematical benchmark functions. In the following, the algorithm's performance on the hypothetical systems is evaluated by performing sensitivity analysis on the effective coefficients of the algorithm, the number of particles, and the number of logical iterations to be implemented on four reservoir systems. The results of the best coefficients showed the proximity of 95.33% to the absolute solution of the four-reservoir system, i.e. 308.29, which implies the good and high accuracy of the algorithm. Then, to supply water downstream of Dez dam while addressing the monthly flow of sediment in the reservoir, which is a very important environmental phenomenon, the atom search algorithm and four other common optimization algorithms are applied, the results of which were prioritized using multi-criteria decision-making (MCDM) process by TOPSIS ranking technique. The results indicated the superiority of the Atom Search Algorithm (ASO), Salp Swarm Algorithm (SSA), Sine Cosine Algorithm (SCA), Particle Swarm Optimization (PSO) algorithm, and Genetic Algorithm (GA), respectively. Therefore, the use of the ASO algorithm in solving optimization problems, especially in the field of reservoir operations concerning environmental factors such as sedimentation is recommended.