The Brazilian state of Pernambuco has approximately 88% of its territory located in the semiarid region, suffering from the irregularities of precipitation and the phenomenon of droughts, which increases the need for constant monitoring of climatic events. This study consisted of estimating meteorological and agricultural drought for the study area with soil moisture data from the Soil Moisture and Ocean Salinity satellite and MODIS products, covering the period from 2010 to 2017 and considering three climatic regions in Pernambuco: Mata, Agreste and Sertão. This research was divided into two main phases: the first was to obtain drought indices calculated from the remote sensing information, with emphasis on producing the Soil Moisture Agricultural Drought Index (SMADI) that uses the soil moisture data in its calculation; the second phase was to correlate the response of the severity of SMADI and other indices with the results of productivity and harvested area for corn and sorghum crops, as obtained from official databases. The results found in the first phase of the study using the SMADI index obtained for the agricultural areas of the state showed long consecutive periods of drought, lasting longer than one year. The hydrological year from 2012 to 2013 presented the longest duration of the event for the Mata and Sertão regions, while for the Agreste the longest period was from 2016 to 2017. The correlations of corn production with the severity of the SMADI values were more significant than sorghum because corn is more sensitive to water stress. The study also concluded that the severity detected by SMADI correlated more significantly with sorghum culture than any of the other indices; that is, SMADI was more sensitive to identify water stress in this crop than the other indices used. The results demonstrated that SMADI can provide early warnings of drought impacts on rainfed agricultural systems and may serve to indicate the adoption of more resistant crops to water stress conditions.