Droughts in Northeast Brazil, which tend to intensify due to climate change, have repeatedly brought famine, mass migration and social conflicts in this region. Its prediction, monitoring and management, however, remain a central research theme. In water resources management in semiarid regions such as the Northeast of Brazil, it is fundamental to have tools to aid decision making. This paper presents three components of the so-called SIGES (Drought Management System), the items related to drought prediction and monitoring, as well as many reservoir operation methodologies for water scarcity situations. Statistical models, artificial neural networks and machine learning techniques were used for drought prediction. In order to perform precipitation monitoring, several indexes were adapted and incorporated into a droughts basic characteristic monitoring system (duration, severity and intensity), so that different mitigating actions could be implemented in accordance with the values reached by these parameters. We utilized the following meteorological indexes for this purpose: Rainfall Anomaly Index (RAI); Bhalme and Mooley Drought Index (BMDI); Lamb Rainfall Departure Index (LRDI). Finally, some reservoir operation for water scarcity situations methodologies are presented and discussed. The described components were applied to the Northeast of Brazil, especially Piauí, Ceará and Rio Grande do Norte states.