A B S T R A C TThe semi-arid region of Northeast Brazil (NEB) is known to be an extremely oscillating region from the climatic point of view, due to the temporal and spatial variability of precipitation. The objective of this work was analyzed the internal seasonal and interannual variability of the precipitations, through dry spells, in the State of Ceará. This behavior is directly influenced by the Sea Surface Temperature (SST) of the Pacific and Atlantic oceans. The RSM97 Regional Spectral Model, through the dynamic downscaling technique, was used to quantify variations using dry spells events (consecutive days with rainfall below 2 mm, occurred during the rainy season of the state), from 1974 to 2012, through FUNCEME daily precipitation data, atmospheric numerical model simulations and SST anomalies (ENSO phases). The results were presented in two stages, first diagnostic mode (with SST anomalies observed from 1974 to 2000) and the second prognostic mode (with persistent SST anomalies from 2002 to 2012). It is concluded that the RSM97 model, predicting drought periods during the rainy season, is reasonable, to a great part the results the model underestimated the number of dry spells in relation to those observed, but represented the signal of the observations of the events which indicates that its physical characteristics influenced the quantification of precipitation and its spatial distribution in the state