Statistical analysis proves that El Nino and La Nina are responsible for up to 40% of annual precipitation variations and up to 30% of river discharge variations in Florida. The analysis is based on 44-year records of precipitation from more than 30 gauge stations and stream discharge from 20 gauge stations distributed all across Florida Peninsula. The cross-correlation coefficients for both the sea surface temperature (SST) and precipitation data series, the SST and river data series are calculated after the SST data series, precipitation and river data series are prewhitened by an autoregressive moving average (ARMA) model (0, 1). The cross-correlations between the SST anomalies and both the precipitation and river discharge are positively significant. The conclusion is that a higher annual precipitation amount (a wet year) is expected from an El Nino year, and a lower precipitation amount (a dry year) is expected from a La Nina year. Large amounts of fresh water recharge into the estuary in an El Nino year and less fresh water recharges into an estuary in a La Nina year. Also a higher groundwater table is expected in an El Nino year, and a lower groundwater table is expected in a La Nina year. Assuming that SST anomalies are the input signals for a time-series analysis, the impulse response weights of both precipitation and river discharge to SST signals can be calculated due to their positive correlations. The impulse response weights can be used to build the linear transfer functions of precipitation, river discharge and SST signals. The annual precipitation and stream discharge amount therefore can be predicted from the SST anomalies. This can provide some guidance for the water management policy and planning.