Evapotranspiration (ET c ) fluxes from plastic mulch production systems are different than open field production systems and require quantification of ET c for plastic mulch environment. A 3-year study was conducted to quantify bi-weekly ET c and develop a crop coefficient (K c ) for drip-irrigated watermelon grown with plastic mulch in sub-tropical Florida using four large drainage lysimeters. The average seasonal ET c was 278mm (min=244mm and max=344mm). Variability in ET c was mainly influenced by rainfall, especially during the initial growth period when it was high. The initial, mid-season, and late season K c values were 0.65, 1.01, and 0.71, respectively. The initial K c was considerably higher than literature values. This finding was mainly due to high soil moisture at the beginning of the growing season resulting from surface application of water for plastic mulched raised bed preparation and additional wetting from rainfall that increased evaporative flux. The K c values were statistically higher (p=0.008) than FAO-56, which underestimated ET c by 30%. A polynomial model was developed to predict K c as a function of days after transplanting. To improve the ET c estimates for the first two bi-weekly periods when the large area of bare and wet soil results in significantly higher evaporation, a multivariate model (r 2 =0.78) was developed to predict the K c as a function of rainfall and relative humidity. The model can be used to adjust K c , and therefore ET c , for the initial stages when evaporation accounts for most of ET c and is sensitive to frequency of wetting. Use of K c values from this study will help improve the accuracy of ET c estimates for drip-irrigated watermelon in subtropical Florida and elsewhere with similar environmental condition.