The seasonality of energy generation sources impacts the operation and planning stages of the electricity sector, as they cause variations in the generation profile throughout the year. Different types of plants, such as hydraulic, wind, and biomass thermal plants, have their availability impacted due to the rainfall regime, the intensity of the winds, or the harvest periods, respectively. In this dissertation, a study is conducted on the impact of hydrological seasonality on the accuracy of network equivalents through probabilistic power flow via Monte Carlo simulation. Reduced networks are widely used in planning studies to operate and expand the electrical system. One of the advantages of its adoption is the possibility of performing a high number of simulations with less computational demand. The precision of the equivalents in this dissertation is quantified, considering the uncertainties in demand, variation of the generation profile, and the occurrence of contingencies of transmission elements. Three network reduction methods are considered in two simulation environments: chronological and non-chronological. The first is to analyze the impact of using an equivalent obtained from the typical configuration of a specific month in the other months of the year. The second aims to compare the different network reduction methodologies in the face of variations in the reduction point, such as transmission contingencies. The studies are carried out with the IEEE 24 bus and IEEE 118 bus systems. The equivalents are created with the Organon software and evaluated in MATLAB. The simulation results are widely discussed, highlighting the importance of using the reduced network to guarantee consistent results.