The South Florida Water Management District (SFWMD) and the U.S. Geological Survey (USGS) have evaluated projections of future droughts for south Florida based on climate model output from the Multivariate Adaptive Constructed Analogs (MACA) downscaled climate dataset from the Coupled Model Intercomparison Project Phase 5 (CMIP5). The MACA dataset includes both Representative Concentration Pathways 4.5 and 8.5 (RCP4.5 and RCP8.5). A Microsoft Excel workbook is provided which tabulates model drought-evaluation statistics for the period 2056-95 based on drought characteristics derived from climate models downscaled by the MACA method assuming the Kruijt stomatal resistance curve in the future. Model drought-evaluation statistics based on 6-mo. and 12-mo. averaged balance anomaly timeseries are provided for four regions: (1) the entire South Florida Water Management District (SFWMD), (2) the Lower West Coast (LWC) water supply region, (3) the Lower East Coast (LEC) water supply region, and (4) the Okeechobee plus (OKEE+) water supply meta-region consisting of Lake Okeechobee (OKEE), the Lower Kissimmee (LKISS), Upper Kissimmee (UKISS), and Upper East Coast (UEC) water supply regions in the SFWMD. The balance anomaly timeseries are computed as the departure from the long-term monthly means of monthly balances (precipitation - reference evapotranspiration) for the period 1950-2005. Then 6-mo. and 12-mo. moving averages of the balance anomalies are computed and drought-event characteristics (duration, intensity and severity) are derived from the moving-average timeseries. The model drought-evaluation statistics include percentile rank of total future (2056-95) drought severity, percentage change in total future drought severity with respect to the historical period (1950-2005) and its percentile rank, as well as p-values comparing the joint distributions of model historical drought characteristics against those derived from observations, and p-values comparing model future to model historical drought characteristics. The lower the p-value, the more different the compared joint distributions of drought characteristics. P-values smaller than a chosen significance level (typically 0.05–0.1) denote models for which the joint distributions of drought characteristics are statistically significantly different.