Eddy covariance (EC) and micro-meteorological data were collected from May 2010 to January 2013 from urban, non-irrigated bahiagrass (Paspalum notatum) in subtropical south Florida. The objectives were to determine monthly crop coefficients (Kc) for non-irrigated bahiagrass by using EC evapotranspiration (ET) data and the Food and Agriculture Organization 56 Penman–Monteith reference evapotranspiration equation; compare crop ET (ETc) calculated with new Kc values to ETc obtained using Kc values available in the literature; and compare results and methodologies for statistical differences. New Kc values ranged from 0.62 to 0.92 and were different from Kc values found in the scientific literature for bahiagrass. Resulting ETc calculated using literature Kc values were significantly different from EC ET data, whereas ETc using the new Kc values was not. Specifically, literature Kc values were temporally biased to miscalculate the timing of convergence between potential and actual ET, assuming that our new Kc values calculated with EC methods were most accurate. As a consequence, ETc calculated using the literature Kc values was either too large or too small. However, one set of literature Kc values from a similar climate and water table depth were closer to our new Kc values, indicating that climate should be considered when selecting urban non-irrigated Kc from the literature to estimate ET. Results also indicated that more than 1 year of EC ET data was needed when establishing monthly Kc values because of annual variability in factors controlling ET, such as water availability. The new Kc values reported herein could be used as an estimate for urban non-irrigated bahiagrass within similar climates. Copyright © 2013 John Wiley & Sons, Ltd.