We present a new framework for detecting and forecasting large spatial extent floods in the Northeastern United States, Hydrological Unit Code 2 (HUC2) region. Historical streamflow data was collected for the years 1955 – 2017 for 55 stations. Floods are classified as days when station’s streamflow exceeds the 99^th percentile of flowrates that were measured during this period. Large spatial extent floods, designated as Spatial Extreme Flood Events (SEFEs), are defined as days with simultaneous floods in more than 3 stations during the same 1955 – 2017 period. The applied methodology resulted in 1,935 SEFEs with a dominant season of January – May (JFMAM). Using field correlation analysis, we found that El Nino Southern Oscillation (ENSO), Pacific Decadal Oscillation (PDO), and northeastern Caribbean sea-surface temperatures (SSTs) have a statistically significant correlation up to 1 season ahead with SEFEs. Hence, these pre-season climate predictors were used to forecast the number of SEFEs in JFMAM. Preliminary modeling using a nonparametric k-nearest neighbor (k-NN) approach reveals promising results with good predictive abilities in a real-time forecasting mode. These forecasts can then be used for quantifying the spatial extent and risks to prepare for water management and emergency services ahead of the season.